Category Archives: Ai

Home to some of the worlds top artificial intelligence (AI) labs, Europe is brimming with AI experts and many of them are using Twitter to talk about their work.

Theyre sharing academic papers, job opportunities, AI news and other bits of information that can be hard to come by through a standard Google search.

But finding these people on Twitter isnt always easy, so Sifted has curated a handy list of AI gurus for you to follow and thrive off.

Neil Lawrence is a researcher at the University of Cambridges Department of Computer Science and Technology. He announced he was joining the university in September as the inaugural DeepMind professor of machine learning. That essentially means DeepMind is paying him to carry out AI research at the university.

Get the Sifted Newsletter

Thank you for subscribing to the newsletter!

Prior to taking up the professorship, Lawrence was director of machine learning at Amazon Cambridge.

Likes to tweet about: Safe AI, Cambridge, diversity

Twitter handle: @lawrennd

It feels like this man needs no introduction, but for anyone who doesnt know who Demis Hassabis is, heres the lowdown. Hes the cofounder and chief executive of the London-headquartered DeepMind AI lab, which was acquired by Google in 2014 for 400m. Prior to DeepMind, Hassabis had his own computer games company called Elixir Studios, but his passion for games goes way back. He was a chess master at the age of 13 and the second-highest-rated under 14 player in the world at one time.

Likes to tweet about: Science, games, DeepMind breakthroughs

Twitter handle: @demishassabis

Catherine Breslin is a machine learning scientist and consultant based in Cambridge. After completing her PhD at the University of Cambridge, Catherine went on to work on automatic speech recognition, natural language understanding and human-computer dialogue systems for the likes of Toshiba Research and the Amazon Alexa team.

Likes to tweet about: AI pioneers, jobs, education, Amazon

Twitter handle: @catherinebuk

Not one to shy away from complex world issues, Berlin-based Samim Winiger recently worked for Google on interactive machine learning projects. Hes also the cofounder of a startup called Ecospace, which is on a mission to cultivate nature with digital tools and practices that are open-sourced to everyone.

Likes to tweet about: Mental health, Japanese culture, inequality, harmful AI

Twitter handle: @samim

Beth Singler is a research fellow anthropologist at the University of Cambridge who, according to her Twitter bio, is thinking about how you think about AI/robots. Among her main concerns are the social, ethical, philosophical and religious implications of advances in AI and robotics.

Likes to Tweet about: Crazy robots, memes, tea, religion

Twitter handle: @BVLSingler

Odette Scharenborg is an associate professor at Delft University of Technology, where she works on speech processing. Shes interested in areas like non-nativeness, under-resourced languages, background noise, and emotion.

Likes to tweet about: Diversity in AI, baby animals, fantasy novels, running

Twitter handle: @OScharenborg

Andrew Trask is a PhD student at the University of Oxford and a senior research scientist at DeepMind, where he studies privacy and AI. Hes also written a book about deep learning, an AI technique, which has sold over 10,000 copies.

Likes to tweet about: Privacy, deep learning

Twitter handle: @iamtrask

Jack Kelly left DeepMind (not many people do) in order to set up a non-profit climate change startup.Open Climate Fix is entirely focused on using open-science to mitigate climate change, Kelly told Sifted earlier this month. The aim of our first project is to reduce emissions from the electricity system by building the best near-term solar electricity forecasting system. Were using machine learning, satellite imagery and numerical weather predictions.

Likes to tweet about: Emissions, startups, National Grid

Twitter handle: @jack_kelly

Nando de Freitas sold his startup, Dark Blue Labs (a spinout from Oxford University), to DeepMind in October 2014. In his Twitter bio, he writes: I research intelligence to understand what we are, and to harness it wisely.

Likes to tweet about: The brain, climate change, DeepMind research

Twitter handle: @NandoDF

Danielle Belgrave is a computer scientist at Microsoft Research. Among other things, she focuses on how machine learning can be applied to healthcare. For example, shes currently looking at using machine learning models to understand disease progression and heterogeneity disease.

Likes to tweet about: Healthcare, climate change, automation

Twitter handle: @DaniCMBelg

Director of AI at PwC, Rob McCargow certainly breaks the mould of a boring accountant working at one of the Big Four. A keen retweeter/news sharer, McCargow does a good job of keeping on top of AI news and sharing it with his followers.

Likes to tweet about: Responsible AI, veganism, UK tech

Twitter handle: @RobMcCargow

Edward Grefenstette left DeepMind in November 2018 and joined Facebook AI Research in January 2019, raising a few eyebrows in the process. One of his main areas of interest is natural language understanding, which has wide applications across Facebooks platform.

Likes to tweet about: Politics, AI research papers

Twitter handle: @egrefen

James Vincent is one of the best-known AI journalists in the UK. Funny, yet smart, he has a knack for sifting through complex AI research papers and pulling out the most interesting bits. Hes also working on a book on the history of measurement.

Likes to tweet about: AI breakthroughs, scary robots, science, weird things on the internet

Twitter handle: @jjvincent

William Tunstall-Pedoe is responsible for building a lot of the technology inside one of the worlds most popular voice assistants: Alexa. He sold his voice recognition startup, Evi Technologies, to Amazon in 2012 for what wasreported to be around $26 million (21 million). Amazon used Evi as the foundation for Amazon Cambridge, which now employs hundreds of people.

Likes to tweet about: Politics, startups, Alexa

Twitter handle: @williamtp

Murray Shanahan splits his time across the Department of Computing at Imperial College London, one of the leading universities in the world, and DeepMind, where he is a senior research scientist. Hes also written a book called The Technological Singularity.

Likes to tweet about: Robotics, computational neuroscience, and philosophy of mind

Twitter handle: @mpshanahan

Verena Riser is a professor in conversational AI at Heriot Watt University in Edinburgh. Her particular area of interest is at the intersection of language technology and machine learning. She has worked on interdisciplinary applications spanning decision support, human-robot interaction, spoken dialogue systems and natural language generation.

Likes to tweet about: Academia, feminism

Twitter handle: @verena_rieser

Joanna Bryson founded the Bath AI Group at the University of Bath. She is set to become a professor of ethics and technology at Berlins Hertie School of Governance in February.

Likes to tweet about: AI ethics and policy

Twitter handle: @j2bryson

Best known for his appearances on Robot Wars, Noel Sharkey is also a professor of AI and robotics at the University of Sheffield.

Likes to tweet about: Biologically inspired robotics, cognitive processes, history of automata

Twitter handle: @NoelSharkey

Tabitha Goldstaub is the brains behind CognitionX, the UKs biggest AI conference. The event attracts employees from companies like Amazon and Palantir, as well as investors looking to back the next big thing. Goldstaub is also chair of the UK Governments AI Council.

Likes to tweet about: AI conferences, women in tech

Twitter handle: @tabithagold

Alan Winfield is a professor of robotics ethics at UWE Bristol, where his work spans research and public engagement. He spends a chunk of his time at the Bristol Robotics Lab and has a particular interest in cognitive robotics.

Likes to tweet about: Politics, robot ethics

Twitter handle: @alan_winfield

John Danaher is an academic and lecturer at the National University of Ireland, Galway. He teaches in the universitys School of Law but hes an affiliate scholar at the Institute for Ethics and Emerging Technologies. His research involves looking at the ethical, social and legal implications of emerging technologies, such as AI.

Likes to tweet about:The mind, surveillance

Twitter handle: @johndanaher

The second PricewaterhouseCoopers (PwC) employee to make the list. Whod have thought it? Maria Axente is the artificial intelligence programme driver and AI for Good Lead at PwC in London. In her role, Axente advises PwCs partners across industry, academia, governments and more on how to harness the power of AI in an ethical and responsible manner.

Likes to tweet about: AI regulation, ethics

Twitter handle: @maria_axente

Sarah Porter is the founder and chief executive of Inspired Minds, which is a community of 52,000 people that are trying to do good with AI.

Likes to tweet about: Climate change, ethics, the EU

Twitter handle: @SColesPorter

Kate Devlin is a senior lecturer in the Department of Digital Humanities at Kings College London, where she is investigating how people interact with and react to technology in order to understand how emerging and future technologies will affect us and the society in which we live. She started her career as an archaeologist, before moving into computer science. Devlin has also written a book on sex robots called Turned On.

Likes to tweet about: Sex robots

Twitter handle: @drkatedevlin

See more here:

30 AI people in Europe to follow on Twitter - Sifted

By Nicole Schmidt

Back in the 70s, renowned mathematician James Lighthill predicted that machines would never be capable of reasoning or even simple tasks, like being able to checkmate a chess pro. Fifty years later, artificial intelligence programs have not only defeated the worlds top chess players, but they can match and in some cases, outperform humans when it comes to art, science and even companionship. Heres a look at nine outrageous things AI can do.

1. Impersonate Vincent van Gogh

Vincent van Goghs painting technique was considered to be so revolutionary that art critics thought it was inimitable, but scientists from Germanys Bethge Lab invented a program that can replicate his iconic brush strokes. The AI system analysed the colours, shadows and highlights of Starry Night, then used the same techniques to create an original painting of the Neckar river in Tuebingen, Germany.

2. Compose classical music

Most creatives will argue that human emotion the one thing computers lack is the key to producing meaningful art. But some companies, like Aiva Technologies, are challenging that notion. The startup created an Artificial Intelligence Virtual Artist that learned music theory by studying a large library of famous composers (including Bach, Beethoven and Mozart). The machine is now capable of creating original classical melodies in a matter of minutes, many of which have been used in films and video games.

3. Make Barbie come to life

Barbies AI-powered doll is something straight out of a Toy Story movie. She has a tiny microphone concealed in her necklace that can analyze speech and respond accordingly (all in less than a second). If the ability to have a coherent conversation wasnt impressive enough, she can also remember what children tell her and refer to it in future exchanges. (Whether the toy falls into the cool or creepy camp is still up for debate.)

4. Be your romantic partner

Harmony AI took notes from Black Mirror to create their AI-powered sex doll. Just as clients can decide whether they want a blonde or brunette, they can also customize their dolls personality. Similar to Hello Barbie (but the R-rated version), the doll uses Bluetooth to hear and respond. Since it comes with built-in memory, its makers say it can develop meaningful relationships with its user over time and it could be all yours for a cool $20,000.

5. Make you dance

Using just a single full-body photograph, researchers at Nvidia, a California-based gaming company, can turn anyone into a backup dancer from one of Beyonces music videos. Their software learns mapping functions from pre-recorded videos and applies those movements to photographs.

6. Turn you into a digital puppet

One of AIs biggest ethical conundrums is the deepfake a fake video or audio recording that looks and sounds like the real deal. Similar to how Nvidia researchers create their dancing videos, the software can essentially perform a virtual head and body transplant by decoding and then reconstructing a brand new face. There is also software that can replicate someones voice, or essentially turn them into a digital puppet. Some of the results (Beck Bennett impersonating a shirtless Vladimir Putin) are hilarious, while others (hyper-realistic revenge porn) are downright terrifying. Back in 2017, a particularly realistic speech delivered by an AI Barack Obama made the rounds on the internet.

7. Detect illness via smell

In a few years, diagnosing an illness could be as easy as taking a breathalyzer test. A team of scientists from around the globe have developed an artificial intelligence system that can detect 17 diseases including Parkinsons, Crohns and several types of cancer using only a breath sample. The NaNose was trained to identify patterns in the chemical makeup of specific diseases by analyzing more than 8,000 patients scents. Right now, the system has an 86 per cent accuracy rate.

8. Read your mind

In whats been hailed as the last privacy frontier, Facebook is working on creating a device that can read your mind. The real life veritaserum (for the non-Harry Potter fans, thats truth serum) can decode brain signals and translate them into words. The short-term goal is to use the technology to help patients with paralysis, but eventually, Facebook wants to make a wearable headset available to the masses that would enable users to control digital devices with their minds.

9. Find missing children

Last year, police in New Delhi used a new facial recognition software to capture images of 45,000 kids around the city, then find matches in Indias database of missing and vulnerable children. They were able to identify 3,000 missing children in just four days.

Continue reading here:

Nine outrageous things AI can already do including compose music and make you dance - National Post

Along with artificial intelligence (AI) and machine learning, many discuss virtual reality (VR) and augmented reality (AR) as the next technologies that will transform how we live our lives and how businesses operate globally. And one of the only ways to stay relevant today is to keep up with these swift changes sweeping across industries and our daily lives.

At my company, we have a few VR companies in our portfolio and have worked on their go-to-market strategies and infrastructure; I have seen this type of technology gain traction exponentially, open new doors and offer new possibilities to businesses and individual users alike.The merging of AI and VR is more pervasive than it seems. Think of how many times you have seen people pose in front of their smartphone cameras and try different filters on Snapchat, Instagram and Facebook. Although this is a very simple and innocent application of the technology in itself, its a great example of how the technology has penetrated and gotten seamlessly integrated in every moment of our lives -- even if it comes in the shape of doggy ears or cat whiskers.

The effect of AI and VR doesnt end there, though; it has the potential to unleash incredible opportunities on a higher level. A vivid example of how AI and VR can transform science -- medicine in particular -- is a machine learning-powered microscope. Researchers on the project developed algorithms to detect breast cancer metastases in lymph nodes with a high level of accuracy; if successful, it could offer pathologists the possibility to accelerate and improve the process of adoption of such progressive technologies for global patient treatment. Technologies like these could offer a chance to increase the availability of high-quality healthcare to patients around the world.

The AI, AR and VR revolution hasnt left the learning and workplace training industry untouched, either.

Google Expeditions is a technology that brings the chance to explore the world from one single classroom using AR and VR. According to the Google website, expeditions include explorative content on history, science, the arts and the natural world. Whether the students study dinosaurs in their natural habitat or wander around admiring Renaissance art, the technology offers experiences with a wide range of subjects.

Microsoft has tapped into VR and AR possibilities, too, with its HoloLens 2 -- an immersive mixed reality technology that, among other things, is designed to help increase student engagement and retention through 3-D technologies. The technology has been applied for workplace training as well: According to The New York Times, Microsofts largest known HoloLens customer (paywall) is the military. The Pentagon reportedly invested $479 million in a contract with Microsoft to provide technology for soldier training on the battlefield to increase skills like mobility and situational awareness.

Other in-office VR applications aim to boost productivity and streamline workflow. Spatial, for instance, can turn any room into a VR-powered shared workspace that remote users can use to collaborate, brainstorm and exchange content. Technologies like these make it possible to share and organize 3-D models, videos, documents, images, and websites through VR headsets.

AI, AR and VR applications have transformed the relationships between customers and brands. These technologies can open the doors for extra personalization and customization of companies' products, content and messages to the specific preferences and needs of target audiences.

Take, for example, Volvo and its initiative to capture the showroom experience in mixed reality. The automobile company implemented VR to capture the experience of viewing a new car in the showroom, and also created an app that allows people to take a car for a test drive in 3-D video models in such a way that potential buyers can experience it from the comfort of their living room.

Even the film industry -- namely, the popular TV show Game of Thrones -- has used VR experiences. The Ascend the Wall VR experience transported the audience into the show's scenery. The final product introduced and manifested a whole new way of storytelling and user engagement and demonstrated how traditional brand marketing can be delivered in innovative, technology-driven ways.

You may be wondering if VR and AR technologies can truly generate a positive impact in the corporate space.

Well, according to 2018 research by Capgemini (via PR Newswire), "82% of companies that are currently in the process of implementing AR and VR solutions say the benefits are either meeting or exceeding their expectations." Additionally, 50% of surveyed executives whose companies werent implementing AR and VR as of today will begin exploring them for their business operations within the next three years.

This all sounds promising, but before taking steps to acquire AR and VR technologies for your company, there are a few questions to consider to ensure a smooth transition to a more digitized company.

Consider what the best use cases for such technology are. Do you need to enhance user experiences with your brand? Or do you need to streamline the workflow between employees scattered around the globe? Or maybe your goal is to boost collaboration by allowing your employees to easily visualize their work. In any of these cases (and many others), you have to consider if and where the technology would yield the greatest impact; dont get blinded by the shine and sparkle of the new technology.

And most importantly, listen to your employees. Make sure they are ready for the transition, understand the benefits of it and embrace the change. This requires working with them to understand where the technology can generate the most positive change and where they could leverage the technology themselves to accelerate the companys growth. The technology should be there to maximize the talent and loyalty of your employees, not to replace them or diminish their importance.

View original post here:

The Impact Of VR, AI And AR In The Workplace - Forbes

Artificial intelligence (AI) -- the capability of a machine to mimic human thinking and behavior -- is one of the biggest growth trends today.Spending on AI systems will increase by more than two and a half times between 2019 and 2023, from $37.5 billion to $97.9 billion, for a compound annual growth rate of 28.4%,according to estimates by research firm IDC. Other sources are projecting even more torrid growth rates.

There are two broad ways you can get exposure to the AI space:

With this background in mind, let's look at which AI stocks are performing the best so far this year (through Nov. 25) and which one is my choice for best AI stock for 2020.

Image source: Getty Images.

The following chart isn't meant to be all-inclusive, as that would be impossible, and the chart has limits on the number of metrics. Notable among the companies missing areAdvanced Micro Devices and Intel. They were left out largely because NVIDIA is currently the leader in supplying AI chips. While there are things to like about shares of both of these companies, NVIDIA stock is the better play on AI, in my view.

Data by YCharts.

Graphics processing unit (GPU) specialist NVIDIA (NASDAQ:NVDA), e-commerce and cloud computing service titanAmazon, computer software and cloud computer service giant Microsoft, Google parent and cloud computing service provider Alphabet, old technology guard and multifaceted AI player IBM, and Micron Technology, which makes computer memory chips and related storage products, would best be put in the first category above. They produce and sell AI-related products and/or services. They're all also probably using AI internally, with Amazon and Alphabet being notably heavy users of the tech to improve their products.

iPhone makerApple (NASDAQ:AAPL), social media leader Facebook (NASDAQ:FB), video-streaming king Netflix, and Stitch Fix, an online personal styling service provider, would best be categorized in the second group since they're either primarily or solely using AI to improve their products and services.

Now let's look at some basic stats for the three best performers of this group.

Company

Market Cap

P/E(Forward)

Wall Street's 5-Year Estimated Average Annual EPS Growth

5-Year Stock Return

Apple

NVIDIA

Facebook

S&P 500

--

--

Data sources: YCharts (returns) and Yahoo! Finance (all else). P/E = price-to-earnings ratio. EPS = earnings per share. Data as of Nov. 25, 2019.

On a valuation basis alone, Facebook stock looks the most compelling when we take earnings growth estimates into account. Then would come Apple and then NVIDIA. However, there are other factors to consider, with the biggie being that projected earnings growth is just that, projected.

There's a good argument to be made that NVIDIA has a great shot at exceeding analysts' earnings estimates. Why? Because it has a fantastic record of doing so, and all one needs to do is listen to enough quarterly earnings calls with Wall Street analysts to realize why this is so: A fair number of them don't seem to have a strong grasp of the company's operations and products. (I'm not knocking, as most analysts don't have technical backgrounds, and they cover a lot of companies.)

Facebook stock probably has the potential to continue to be a long-term winner. But it's relatively high regulatory risk profile makes it not a good fit for all investors. Moreover, it will likely have to keep spending a ton of money to help prevent "bad actors" from using its site for various nefarious purposes. Indeed, this is one of the major internal functions for which the company is using AI. It also uses the tech to recognize and tag uploaded images, among other things.

Apple uses AI internally in various ways, with the most consumer-facing one being powering its voice assistant Siri. It's the best of these three stocks for more conservative investors, as it has a great long-term track record and pays a modest dividend.NVIDIA, however, is probably the better choice for growth-oriented investors who are comfortable with a moderate risk level.

Image source: Getty Images.

NVIDIA is the leading supplier of graphics cards for computing gaming, with AMD a relatively distant second. In the last several years, it's transformed itself into a major AI player, or more specifically, a force to be reckoned with in the fast-growing deep-learning category of AI. Its GPUs are the gold standard for AI training in data centers, and it's now making inroads into AI inferencing. (Inferencing involves a machine or device applying what it's learned in its training to new data. It can be done in data centers or "at the edge" -- meaning at the location of the machine or device that's collecting the data.)

NVIDIA is in the relatively early stages of profiting from many gigantic growth trends, including AI, esports, driverless vehicles, virtual reality (VR), smart cities, drones, and more. (There is some overlap in these categories, as AI is involved to some degree in most of NVIDIA's products.) There are no pure plays on AI, to my knowledge, but NVIDIA would probably come the closest.

Read the original:

The Best Artificial Intelligence Stocks of 2019 -- and The Top AI Stock for 2020 - The Motley Fool

In the face of rising demand for data, increasingly saturated mobile markets, and stiff opposition from legacy players, tech entrants, and startups, global telecoms are locked in a battle for market share. These market pressures have led to vicious price wars for mobile services and, as a result, declining average revenue per user (ARPU).

Making matters worse, improvements in infrastructure and technology have made telecoms largely comparable in terms of coverage, connection speeds, and service pricing, meaning companies must transform their businesses if they hope to compete.

For many global telecoms, shoring up market share under today's pressures while also future-proofing operations means having to invest in AI. The telecom industry is expected to invest $36.7 billion annually in AI software, hardware, and services by 2025, according to Tractica.

In the AI in Telecommunications report, Business Insider Intelligence will focus on the use of AI to enhance the customer experience, which can directly impact revenue. Each year, an estimated $62 billion is lost by US businesses after inferior customer experiences, according to NewVoiceMedia. We will discuss the forces driving firms to AI, pinpoint some of the top use cases of AI along the customer journey, and identify some of the leading companies in the space

The companies mentioned in this report are: AT&T, CenturyLink, China Mobile, IBM, Spectrum, Sprint, Swisscom, Telia, T-Mobile, and Vodafone.

Here are some of the key takeaways from the report:

In full, the report:

Interested in getting the full report? Here are three ways to access it:

Read the rest here:

AI IN TELECOMMUNICATIONS: Why carriers could lose out if they don't adopt AI fast - and where they can make th - Business Insider India

Earlier this month, Nasdaq announced that a new deep learning system, aided by human analysts, would be used to watch for fraud in the 17.5 million trades done daily on the stock exchange, the worlds largest in terms of volume.

The announcement marked the latest notable development in a growing area of artificial intelligence (AI) and machine learning (ML) application: prevention of financial fraud. Nasdaqs large number of transactions that is, data and the multiple avenues for fraud those transactions present make it an obvious place to put the best technology has to offer to work. But AI and ML are being used to watch for signs of financial fraud and the other crimes that often accompany that fraud, from credit unions on up.

Fraud detection is one of the classic applications of AI. And it's incredibly important to a lot of industries, said Chris Nicholson, CEO of Skymind. Payment processors like PayPal or Stripe only survive because they have excellent fraud detection. Big banks and insurance companies are in the same position: They either get good at fraud detection or they get robbed.

The changing face of financial fraud is leading to growth for businesses designed to fight it. The fraud detection and prevention market in North America is expected to expand by more than 25% between 2019 and 2027, according to a new report from Absolute Markets Insights.

Companies with a lot of wealth both the data kind and the financial kind can build the AI and ML teams they need to create custom AI and ML solutions. Other businesses rely on other companies such as Verafin, which supplies fraud detection software to credit unions and smaller banks across North America, or SAS, which works with various clients in banking, insurance, government and healthcare.

One of the main ways that enterprise uses AI to detect fraud is through anomaly detection, Nicholson said. That is, they look for unusual behavior because that often correlates to bad behavior.

This has one key advantage in that you dont need a labeled data set of fraud to recognize activity that is merely unusual based on the norm, Nicholson said.

Another thing enterprise will do is try to combine two predictions: what is fraudulent and what is severe enough to be worth going after, Nicholson said. Sometimes fraudulent insurance claims would cost more to fight than to pay out, for example. This is where AI can quickly do a cost-benefit prediction to save a company money, taking fraud detection beyond merely finding the activity, he said.

At HackNotice, personalized recommendations via machine learning provided to customers after a data breach or fraud help reduce the damage from these events, said Steve Thomas, the companys CEO.

We collect every source of information that we can about a data incident, including official state breach disclosures sometimes several different disclosures from different state governments and news articles, Thomas said. The company then maps each type of information to specific threats or actions hackers could take to exploit the information, take over accounts or commit further fraud, and also maps recover actions that can be taken against those threats.

Machine learning is key to this individualized approach, but AI has fraud prevention benefits for enterprises as a whole as well when it comes to reducing threats introduced by their own employees, Thomas said.

Employees are a major security risk for organizations because it is nearly impossible to regulate their activity, and oftentimes security teams are not aware of an employee-caused security event until lasting damage has already been done, Thomas said. Through continuous monitoring of employee accounts and assets, the machine learning algorithm can be trained to find the exposed information and recommend ways to solve the problem.

More here:

Role of AI and ML in Financial Fraud Prevention Grows - ITPro Today

As pressureson healthcare systems intensify, an increasing number of consumers are turning to the use of symptom checkers in the search for fast answers or guidance to any concerns that they may have. But although these tools have become popular, questions around their accuracy, and not only, are plaguing the digital health space.

Last week, at Slush, Ada Health cofounder and chief medical officer Claire Novorol spoke to Wired UKs Victoria Turk about the Berlin-headquarteredcompanys approach to building trust in its AI-powered chatbot.

Absolutely key, first of all, is the quality of the product, Novorol told the audience. So what weve always focused on from the very beginning, eight years ago now actually, is the quality of the core product, the foundation of everything that we do, and thats our knowledge base, our reasoning engine, and how it works, the clinical quality of that, accuracy, safety.

Putting the patient at the heart of everything and being open, transparent and proactive in interactions with all stakeholders are equally as important, Novorol continued.

Initially, Ada Health sought to create a tool that would support doctors in the diagnosticdecision-making process, following Novorol'sexperience of working in the UK's NHS as a paediatrician and as a clinical geneticist, where she was seeing patients with rare conditions.

The team, however,soon identified an opportunity to reach the individual directly. But that required a new approach. We had to make this switch over from being a very clinically-driven, medical-facing product and also launching an app thats available to the everyday individual, the cofounder said.

About four years ago now, we did a lot of work on how to build an interface that was friendly, yet at the same time authoritative and trustworthy, and using very strong clinically, but at the same time very simple, easy to understand, patient-friendly language.

At the moment, the companyisleading a stream within an initiative from the World Health Organisation on benchmarking AI in healthcare, which looks at AI-powered symptom assessments.

We have, as part of that group, academics, industry stakeholders, experts, and a lot of other companies working in our space, our competitors, we work closely with them on this topic of what does good look like, how do we measure accuracy and safety, how do we benchmark these products, Novorolexplained.

However, Ada Health hasnt chosen a pure machine learning approach, according to the CMO, with human experts playing a key role in ensuring that the app and the information provided is continuously improved, feedback from users is implemented and bias prevented from the ground up.

Now, their app isavailablein a variety of languages, including Portuguese, Spanish, and, most recently, Swahili, following a partnership with Fondation Botnar. This was prompted by the team findingthat individuals were more likely to use the tool again if it was available in their own language.

But ensuring thatservicesare tailored to the needs of usersis not just about that. Its making sure that were adapting for local conditions, prevalences of diseases in that [specific] country, then cultural nuances, how you interact with an individual, Novorol said. We work with doctors (), local clinicians trained in that language.

As companies like Ada Health continue to expand, many have been left wondering where this type of digital health toolsfit into the wider healthcare system.Novorol cautioned thattheirapp should be seen as the first port of call, and not a GP replacement. That is because it does not provide a diagnosis or a single condition as the cause of symptoms that someone may experience, but instead offers a range of possibilities and guides the user to seek appropriate care.

Whats very important for us is being very clear about what Ada does and doesnt do, being very clear about the promise and not overpromising,Novorol told the audienceat Slush. The combination of Ada and doctor is powerful.

More:

Ada Health's chief medical officer on AI and building trust in digital health tools - MobiHealthNews

It is worth noting that ACM CIKM 2019 is the 28th session of CIKM and the second time CIKM has come to China since it was first held in 1992. The event gathered many big names. Academician Hong Mei and Ramamohanarao Kotagiri served as the Honorary Chairmen of the conference; Professor Wenwu Zhu and Professor Dacheng Tao served as the Chairmen of the conference; Dr. Wei Cui, the Co-founder and Chief Scientist of Squirrel AI Learning served as the Regional Chairman of the conference, bringing a brilliant sharing of application practice to everyone.

Data mining is the advancement of data analysis, while AI is the extension of business intelligence. In this conference, in addition to Dr. Wei Cui, famous scholar Steve Maybank, Jiawei Han, Jian Pei and Jianping Shi also respectively delivered keynote speeches to explore the development trend of technology. In addition to the necessary tutorial, workshop, oral and poster presentations for the summit, this conference also held the AnalytiCup and more than ten industrial plenary speeches.

Encounter on the Same Field to Explore the In-depth Application of AI Technology

"We believe that in the future, AI will penetrate into all aspects of life. At present, it is mainly applied to deep learning. However, this conference hopes to discuss what kind of development trend will be in the future for deep learning and traditional statistical learning."

As an internationally renowned scholar in the field of AI and information science, Dr. Dacheng Tao, General Chair of the conference, said at the beginning of the conference that many advanced technologies have been understood and mastered by the Chinese people, and even breakthroughs have been made. Under the current trend of science and technology, it is particularly important to explore the in-depth application of new technologies.

Dr. Wei Cui, Co-founder and Chief Scientist of Squirrel AI Learning, also fully agreeswith this. As the first company that has developed the AI adaptive learning engine with complete independent intellectual property rights and advanced algorithm as the core, Squirrel AI Learning has used a variety of AI technologies such as evolutionary algorithm, neural network technology, machine learning, graph theory and Bayesian Network to recommend personalized learning solutions to students in the past few years of practice. The further in-depth application of technologies and the real-time improvement and update of products are closely related to the education status and future of hundreds of millions of students.

At the event, Prof. Jiawei Han, Professor of the Department of Computer Science at the University of Illinois Urbana Champaign, ACM and IEEE Fellow, Director of the American Information Network Academic Research Center, Prof. Steve Maybank, Professor of the Department of Computer Science at the Birkbeck College, University of London, Academician of the European Academy of Sciences and IEEE Fellow, and Wenwu Zhu, Deputy Director of the Department of Computer Science at Tsinghua University, and Deputy Director of the National Research Center for Information Science and Technology and other experts and scholars shared their insights in the field of AI technology with Dr. Wei Cui, the Co-founder and Chief Scientist of Squirrel AI Learning, reflecting the latest research trends in technology.

For example, Professor Jiawei Han noted that it is still a long way to transform real data to structured data and then to useful knowledge. "In order to transform the existing unstructured big data into useful knowledge, the first thing to do is to structure the data. He proposed two forms of structured data, one is Heterogeneous Network, and the other is Multi-dimensional Text Cube. Knowledge generated from this structured data has proven to be very powerful, but it is very difficult to transform the original unstructured data into structured data (Network or Text Cube)."

According to Professor Han, a large number of text data itself implies a large number of hidden patterns, structures and knowledge, and therefore we can use the domain-independent and domain-dependent knowledge base to explore how to transform massive data from unstructured data into useful knowledge. "On this road, we have only found a few openings to go forward. Now it is not a road yet, but just a small path. If we want it to be a broad road, we need to work together. As long as this road is wide, we can transform a large number of unstructured texts into a lot of useful knowledge in the future."

Realize Breakthroughs and Solve the Difficulties in Traditional Education with AI

From industrial production, to family life, and then to voice assistants on various electronic devices, AI and robotics are rapidly transforming this era that belongs to us. For many years, the shortage of senior teacher resources and geographical problems in China's education have affected the popularization of high quality education. But now, AI technology is penetrating into our lives so as to solve the difficulties in traditional education.

Specifically speaking, the intelligent adaptive learning system of Squirrel AI Learning is a student-centered intelligent and personalized education. It applies AI technology to the processes of teaching, learning, assessment, testing and training, to achieve the purpose of surpassinghuman teaching on the basis of simulating excellent teachers. With high cost performance, the product adopts the mode of AI + human teachers to teach students according to their aptitude, which can effectively solve the problems of high class cost, limited famous teacher resources and low learning efficiency of traditional education.

Inhis speech, Dr. Wei Cui introduced the basic techniques used to build the system and performance assessment experiments. Squirrel AI's intelligent adaptive learning engine consists of a three-layer architecture: ontology layer, algorithm layer and interactive system.

The ontology layer mainly focuses on content, including a learning map and a knowledge map. Squirrel AI Learning can split the knowledge points in the discipline into super-nano-level knowledge points, so as to have a clearer understanding on the mastery of knowledge points by students. It can accurately detect the weak points of children's knowledge, and accurately give the most suitable learning path for each child, so as to improve his/her learning efficiency. Taking junior high school mathematics as an example, Squirrel AI Learning can refine 3,000 knowledge points into 30,000.

The algorithm layer includes a content recommendation engine, a student user portrait engine and a target management engine, etc. Squirrel AI Learning will combine the user state assessment engine and knowledge recommendation engine to build a data model, so as to accurately and efficiently detect each student's knowledge loopholes, and recommend corresponding learning content according to the loopholes.

The interactive system is used to learn more about students' information by collecting interactive data and perfect the algorithm. The MIBA multimodal comprehensive behavioral analysis AI system developed by the Squirrel AI Learning and the Stanford Research Institute candetect students' login time, learning time, speed and results, andcapture the children's real-time data, such as eye movements, brain waves and other comprehensive values through monitoringto judge the concentration and focus of students' learning, so as to judge the learning content of the next link.

In addition, the MCM system of Squirrel AI Learning can detect people's model of thinking, learning capacity and learning method. After the assessment and detection, MCM system can analyze different learning capacities, learning speeds, blind spots and weak spots of knowledge points for those learners with the same score, so as to accurately depict the user portrait of learners.

"Compared with the rapid development of Internet technology and AI technology, in fact, our education field is developing very slowly." Dr. Wei Cui pointed out that Squirrel AI Learning is solving the difficulties in traditional education. In the past four years, Squirrel AI Learning has opened more than 2,300 learning centers in more than 700 cities and counties in more than 20 provinces across the country, providing high quality teaching services to nearly 2 million students.

"In the future, we will further strengthen our product technology research and development capabilities, join hands with more experts in adaptive learning to enhance the intelligent adaptive learning system to a higher level."

SOURCE Squirrel AI Learning

See the original post:

Dr. Wei Cui, Chief Scientist of Squirrel AI Learning of Yixue Education, Serves as Local Chair of ACM CIKM 2019 Conference to Explore the Application...

As I read Melanie Mitchells Artificial Intelligence: A Guide for Thinking HumansI found myself recalling John Updikes 1986 novel Rogers Version. One of its characters, Dale, is determined to use a computer to prove the existence of God. Dales search leads him into a mind-bending labyrinth where religious-metaphysical questions overwhelm his beloved technology and leave the poor fellow discombobulated. I sometimes had a similar experience reading Artificial Intelligence. In Mitchells telling, artificial intelligence (AI) raises extraordinary issues that have disquieting implications for humanity. AI isnt for the faint of heart, and neither is this book for nonscientists.

To begin with, artificial intelligence machine thinking, as the author puts it raises a pair of fundamental questions: What is thinking and what is intelligence? Since the end of World War II, scientists, philosophers, and scientist-philosophers (the two have often seemed to merge during the past 75-odd years) have been grappling with those very questions, offering up ideas that seem to engender further questions and profound moral issues. Mitchell, a computer science professor at Portland State University and the author ofComplexity: A Guided Tour, doesnt resolve these questions and issues she as much acknowledges that they are irresolvable at present but provides readers with insightful, common-sense scrutiny of how these and related topics pervade the discipline of artificial intelligence.

Mitchell traces the origin of modern AI research to a 1956 Dartmouth College summer study group: its members included John McCarthy (who was the groups catalyst and coined the term artificial intelligence); Marvin Minsky, who would become a noted artificial intelligence theorist; cognitive scientists Herbert Simon and Allen Newell; and Claude Shannon (the inventor of information theory). Mitchell describes McCarthy, Minsky, Simon, and Newell as the big four pioneers of AI. The study group apparently generated more heat than light, but Mitchell points out that the subjects that McCarthy and his colleagues wished to investigate natural-language processing, neural networks, machine learning, abstract concepts and reasoning, and creativity are still integral to AI research today.

Also read:Artificial Intelligence Cant Think Without Polluting

Mitchells goal is to give a thorough (and I mean thorough) account not only of the ethical issues artificial intelligence raises today (and tomorrow), but of how the various branches of AI that the Dartmouth group pursued actually work. She is a good writer with broad knowledge of the topic (unsurprising, since she has a Ph.D. in computer science), and a canny mindfulness of both the merits and problems of AI. But even so, nonscientists will find it grueling to follow some of her explanations of the technical workings of AI. All too often, I found myself baffled and exasperated when she delved into high-tech arcana.

Take, for instance, the authors discussion of deep learning, which she says is itself one method among many in the field ofmachine learning, a subfield of AI in which machines learn from data or from their own experiences. So far, so good. However, from there matters become tenebrous: Deep learning simply refers to methods for training deep neural networks, which in turn refers to neural networks with more than one hidden layer. Recall that hidden layers are those layers of a neural network between the input and the output. The depth of a network is its number of hidden layers.

From there, she goes, well, deeper, for about eight more pages of text, diagrams, and photos that fail to fully clarify the subject for a general audience. This kind of abstruseness, alas, is fairly frequent, but still, I urge readers to soldier through the technology warrens, because we need to understand the systems that are frightening so many today, and the dedicated reader will come away with at least a modicum of understanding about how AI operates.

I also wish the book had examined the role AI is playing in military weaponry, and how quantum computers affect, or will affect, artificial intelligence or vice versa. In a recent article in the New York Times, Dario Gil, the director of IBM Research, is quoted as saying, The reality is, the future of computing will be a hybrid between [the] classical computer of bits, AI systems, and quantum computing coming together.

Also read: Is This the AI We Should Fear?

The book is exemplary, however, when discussing where AI is now and where it might be going, as well as the moral issues involved. Should we be terrified about AI? she writes. Yes and no. Superintelligent, conscious machines are not on the horizon. The aspects of humanity that we most cherish are not going to be matched by a bag of tricks. At least I dont think so. However, there is a lot to worry about regarding the potential for dangerous and unethical uses of algorithms and data.

Mitchells message is that AI-phobes can chill out because were not now and we probably wont ever be facing a dystopic future controlled by machines. One of the themes of the book is that while its impressive that AI devices have defeated human experts in Jeopardy and Go, no matter how remarkable such tours de force are, those were the only things those particular machines were programmed to do, and they required human input. And in such areas as object recognition, transcribing or translating language, and conversing with Homo sapiens, AI is, to use a word Mitchell favors, brittle.

The hand of humanoid robot AILA (artificial intelligence lightweight android) operates a switchboard during a demonstration by the German research centre for artificial intelligence at the CeBit computer fair in Hanover March, 5, 2013. Credit: Reuters/Fabrizio Bensch

Which is to say that even though great strides have been made (and will be made) in AI, such technology is a long way from being omnipotent, because it is error prone when faced with perplexing to its way of thinking tasks (be cautious, she warns, of riding in self-driving cars). And AI machines are still vulnerable to being manipulated by hackers who might work for foreign governments or are simply motivated to cause mayhem.

Near the end of the book, Mitchell asks, How far are we from creating general human-level AI? She quotes a computer scientist, Andrej Karpathy, who says, We are really, really far away, and then she concurs: Thats my view too.

Above all, her take-home message is that we humans tend to overestimate AI advances and underestimate the complexity of our own intelligence. These supposed limitations of humans are part and parcel of our general intelligence, she writes. The cognitive limitations forced upon us by having bodies that work in the world, along with the emotions and irrational biases that evolved to allow us to function as a social group, and all the other qualities sometimes considered cognitive shortcomings, are in fact precisely what enables us to be generally intelligent.

Also read:Why India Needs a Strategic Artificial Intelligence Vision

It occurred to me while reading about the extraordinary scientists in Artificial Intelligence that as AI becomes more intricately innovative, the men and women working in the field are also keeping pace, also becoming more redoubtably intelligent. So why worry? Surely, if theres ever an AI attempt to subjugate humanity, I have no doubt that Mitchell and others like her, or their successors, will protect our brittle species.

Howard Schneider is a New York City-based writer who reviews books on technology and science. His work has appeared in the Wall Street Journal, the Humanist, Art in America, the American Interest, and other publications.

This article was originally published on Undark. Read the original article.

See more here:

A Guided Tour of AI and the Murky Ethical Issues It Raises - The Wire

As cloud complexity increases, hackers are relying on more targeted attacks, scoping out weak points across a larger attack surface.

Given the proliferation of high-profile attacks in 2019, the security outlook for next yearand the next decadeis filled with potential pitfalls, as challenges persist in maintaining the security profile in enterprises, particularly as security operations teams are spread thinner as attack surfaces widen.

SEE:Special report: The cloud v. data center decision (free PDF)(TechRepublic)

McAfee CTO Steve Grobman and Director of Engineering Liz Maida--who joined the company through their acquisition of Uplevel Security, a firm that applied graph theory and machine learning to security data--spoke to TechRepublic about the security forecast for 2020.

In contrast to spray-and-pray attacks, relying on port scanning to uncover low-hanging vulnerabilities, an increase in attacks targeting specific industries are anticipated to continue their rise in popularity. "We've seen a good number of ransomware campaigns where the adversaries have done reconnaissance to really understand the critical assets [and] the defenses, and then tailor the attack in order to get into that environment, to demand a higher payment from the victim," Grobman said.

"That really requires a much more sophisticated level of defense for the defenders. The other point that I'd make is...we see the evolution of attacks from just focusing on traditional compute environments, to also focusing on cloud environments. Given that many organizations are shifting key components of their operations into the cloud, it would be natural that adversaries are looking for ways to not only target traditional environments, but also cloud assets," Grobman said.

While multicloud deployments may not be optimal from an IT standpoint, large organizationsparticularly those with rich M&A histories, are likely to have footholds in multiple cloud providers. This complicates security measures, though solutions exist to address this deployment scenario.

"A CASB (cloud access security broker) solution allows you to set up a common set of security policies, and have them apply to multiple environments. Similarly, you can do security monitoring and operations at a much lower cost than having to figure out how to instrument each of the cloud environments independently," Grobman said. "You still do have some additional overhead, because you need to have your operators and administrators understand some of the nuanced differences between cloud environments. You end up with some incremental challenges and costs, with every additional provider you add, but at least you can mitigate some of that by using the right technology."

Artificial Intelligence (AI) is, in 2019, the same as blockchain was in 2018every startup wants to bolt it onto their current offering in the hopes of attracting venture capital funding. "You can't just apply machine learning and AI on data that hasn't been set up well to begin with. The actual normalization and understanding that the cleanliness of the data has to be there as a foundational layer, before you can actually start applying different algorithms to extract more intelligence from the information," Maida said.

"One of the common things that we have seen is security analysts trying to correlate events from phishing emails, web gateways, endpoint attachment software, et cetera. If you can't actually understand the data within those events, it becomes very difficult to start understanding, [if] are all of these connected with some unique way, that it would actually suggest the presence of a potential malicious actor," Maida added.

"Cybersecurity is a great example, where you might have a company [claim] their AI algorithm is able to detect 99.999999% of threats. They could show you quantitative data to back up their claim. Unless you know all of the exact right questions to ask, you might look at that technology and say, 'Wow, that's amazing'. If you don't ask questions such as, 'What's your false positive rate required to get that detection rate?" Grobman said. "There's a number of, we can almost think of them as data science sins, that are very easy to abuse in order to make things look better than they actually are."

Strengthen your organization's IT security defenses by keeping abreast of the latest cybersecurity news, solutions, and best practices. Delivered Tuesdays and Thursdays

Image: Welcomia, Getty Images/iStockphoto

Read more here:

Cybersecurity in 2020: More targeted attacks, AI not a prevention panacea - TechRepublic