Category Archives: Ai

Volkswagen is working with Nvidia to expand its usage of its artificial intelligence and deep learning technologies beyond autonomous vehicles and into other areas of business, the two companies revealed today.

VW set up its Munich-based data lab in 2014. Last year it pushed on with the hiring ofProf. Patrick van der Smagt to lead a dedicated AI team that is tasked with taking the technology into areas such as robotic enterprise, or use of the technology in enterprise settings.

Thats the backdrop to todays partnership announcement. VW wants to use AI and deep learning to power new opportunities within its corporate business functions and, more widely, in the field of mobility services. As an example, the German car-maker said it is working on procedures to help optimize traffic flow in cities and urban areas, while it sees the potential forintelligent human-robot collaboration, too.

Artificial intelligence is the key to the digital future of the Volkswagen Group. We want to develop and deploy high-performance AI systems ourselves. This is why we are expanding our expert knowledge required. Cooperation with NVIDIA will be a major step in this direction,Dr. Martin Hofmann, CIO of the Volkswagen Group, said in a statement.

Beyond the work on VWs own brands, the car-maker and Nvidia are teaming up to help other startups in the automotive space. The VW Data Lab is opening a startup support program that is specialized on machine learning and deep learning with Nvidias help. The first batch will include five startups and start this fall. The duo are also reaching out to students with a Summer of Code camp that will begin soon.

Nvidia is already working with VW-owned Audi on self-driving cars which they are aiming to bring to market by 2020 but todays announcement is purely about the data potential and not vehicles themselves. VW did ink an agreement earlier this year to work with Nvidia to develop AI-cockpit services for its 12 automotive brands, but it is also working with rival chip firm Qualcomm on connected cars and smart in-car systems, too.

This VW hookup is one part of a triple dose of automotive-themed news updates from Nvidia today.

Separately, it announced that Volvo andAutoliv have committedto sell self-driving cars powered by its technology by 2021. Nvidia also signed up auto suppliersZF and Hella to build additional safety standards into its autonomous vehicle platform.

See the article here:

Volkswagen partners with Nvidia to expand its use of AI beyond autonomous vehicles - TechCrunch

As AI slowly seeps into various sectors, it was fairly inevitable that we would start to see the AI for X startups begin to appear. Thus AI for small business social media is now a thing, in the shape of PromoRepublic, a US/Finnish/Ukrainian startup which has now raised a $1.2M investment round.

More specifically, they bill themselves as a WIX for a small businesss social presence. So what does that mean? Effectively its a simple way to grow a small business with social media content.

Investors in this seed round include Peter Druckers daughter Cecily Drucker, Nick Bilogorskiy (the ex-chief malware expert at Facebook), angel investor Aviram Jenik, business modeling guru David Lottenbach, Finnish Funding Agency for Innovation Tekes, as well as Howzat Partners, Digital Future and Spring Capital VC firms.

So far they have 50,000 small and medium business owners registered on the platform. It competes with visual design companies like Canva, Stencil, Adobe Spark, and DIFM (do-it-for-me) companies like MainStreenHub, Boostability and RevLocal who do custom social content for SMBs. DIFM-companies tend to cost around $300 per month which is pricey for small businesses. PromoRepublic starts at $20 per month.

The platform helps small business owners understand what specific content to post for their specific business. So, when to post, how often, what copy, design, call to actions and hashtags to use on Facebook, Twitter, LinkedIn and Instagram. Then the user needs to make a final customization and approve content for the whole month. The platform has access to 100,000+ templates and images and is integrated with Hootsuite, HubSpot, Buffer, and Yext.

PromoRepublic will post automatically at the right time and frequency, thus doing the heavy lifting for SME/SMB business owners who really wouldnt know where to begin in terms of creating and sharing content about their business.

Given that most small business owners usually do their own social media, theyd probably happily drop $20 a month on a platform that came up with ideas they could use and approve every month. Goodbye social media consultants? This could well be the case, at least at this level.

More here:

PromoRepublic raises $1.2M for AI which creates social media for small businesses - TechCrunch

Chip Register. Credit: Publicis.Sapient

Two names were uttered more than others at Cannes last week. One is Arthur Sadoun, the new Publicis Groupe CEO who unexpectedly announced that his agency holding company will skip Cannes en masse next year. The other is Marcel, an AI system whose development Publicis will fund with the savings.

But there's another player in the drama, Chip Register, co-CEO of Publicis.Sapient and the architect of Marcel.

Register is nonplussed by the reaction to the announcement, which has included trolling by rival agencies on Twitter and sneering that Marcel is nothing more than an amped-up Alexa or publicity stunt executed by a newbie CEO trying to improve the bottom line.

"I expect and expected the skeptics," said Register, who works out of Arlington, Va., but lives in New Orleans. "That is always the case whenever you've got the idea and nerve to step out like this. My only comment to them is, 'See you at VivaTech.'"

VivaTech is Publicis' annual technology conference in Paris and where the company plans to debut Marcel next year.

Here's how Register describes what Marcel, named after Publicis founder Marcel Bleustein-Blanchet, will be able to ferret out. "In a group of 80,000 people that have 200 capabilities across 130 countries, where is the best talent to work on a project once you've received an RFP or a brief?" he said. "Where is the absolute best talent in the group to work on that and how can we assemble that team and allow that team to work and collaborate virtually to bring the best ideas and values we can to a client at a moment's notice?"

Marcel is "a transformation to go from a group to a platform," he said. "A bunch of organizations to a flat leveling of capability that can be compiled in new creative ways that can solve new and creative problems for our clients. What Marcel does is create the mechanism for that happen."

Register said Marcel will not be an enemy of creativity, but will facilitate it.

"There's been all sorts of speculators and commentators out there saying there's a trade-off between creativity and technology," Register said. "That is an absurd notion."

"The use of technology enables great creative work. It enables the connectivity of people," he added. "It enables teams to work and it enables ideas to generate, be shared globally and virtually, through the use of better insight in culture and the journey of human beings."

Publicis turned to Register due in large part to his role with Sapient before it became part of the holding company in a $3.7 billion acquisition in 2015. His expertise is building tech using Sapient's Global Distributive Delivery system, to which Publicis attributes Sapient's 32% growth rate from 2004 through 2007. Publicis.Sapient has nearly 23,000 employees, more than half of whom run that system from India and will play a large part in the development of Marcel. That's one secret of Marcel's deployment.

"We work in a very virtual way," Register said of Global Distributive Delivery. "It takes a project and it divides the requirements into the places across the whole world where the greatest talent exists to solve those problems."

Sadoun himself was in India a few weeks ago touring tech facilities operated by Register's team. It was there, insiders claim, that Sadoun's idea to ditch Cannes for Marcel was born.

Register and Publicis vigorously refute that.

"I'm not sure I can pin down a moment or an event that led to the idea," Register said. "We've been talking about how to do this for a while."

Register said he and key leaders met after Sadoun made his announcement in Cannes, adding that they "ideated 15 to 20 core competencies for the platform."

Marcel is being built internally because no one can understand the unique customization required to get the most out of its talent base other than Publicis, Register said. The company will likely work with a third-party platform in some capacity, though, to aid in the rollout.

"We buy lots of software from lots of companies that could play a role in the ultimate architecture of the products," Register said. "That's a foregone conclusion."

"But there is no off-the-shelf solution that is going to explode the value of Publicis Groupe," he added. "Fortunately, we are able to do that ourselves because we have a huge technology based enterprise that exists on a wide, global scale. There's a difference between being able to do Einstein's math and being able to split an atom; one is the ability to understand a problem and the other is ability to execute. And that is where I think we have a great shot at leading the transformation of our company."

FIVE THINGS YOU'LL BE ABLE TO ASK MARCEL

1. "Marcel, who is the CMO of Tesla and is anyone in the network connected to him or her? Please also check LinkedIn relationships."

2. "Marcel, do we have any Mumbai-based full-time, temporary, or contract employees with 5 to 7 years Java angular development experience?"

3. "Marcel, can you show me examples of great creative work we have done for luxury apparel clients?"

4. "Marcel, who won awards for creativity from our LA office?"

5. "Marcel, can you help me find a creative director in Chicago with healthcare experience?"

~ ~ ~ CORRECTION: An earlier version of this article misidentified the Publicis.Sapient system that will help develop Marcel. It is Global Distributive Delivery, not Global Distribution Delivery. The article also said Publicis.Sapient has 12,000 employees; the correct figure is 23,000.

Excerpt from:

The Man Behind Marcel, Publicis Groupe's New AI Platform, Expected the Skeptics - AdAge.com (blog)

Wall Street Journal (subscription)

How AI Is Transforming Drug Creation Wall Street Journal (subscription) In the past, drug companies have used artificial intelligence to examine chemistrywhether a drug might bind to a particular protein, for instance. But now the trend is to use AI to probe biological systems to get clues about how a drug might affect a ...

Read more:

How AI Is Transforming Drug Creation - Wall Street Journal (subscription)

The idea of artificial intelligence (AI) has been around for decades, and from movies to games, has now found a home in the mainstream. We are not yet at the point where computers are starting to plot the overthrow of humanity, but through AI, computers are beginning to understand our fellow human beings better than we may understand ourselves.

At least, advertising and marketing platforms that integrate AI into their processes are crunching numbers much faster than we ever could, and deriving insights that marketers use to better target and understand the end user and consumer.

Starting simple with AI technologies, there are recommendation engines: "Early low-hanging fruit for brands to harness the power of AI is in content discovery, Glenn Hower, senior analyst at Parks Associates, told attendees at the "A.I. Meets Media: Innovation Summit" presented by Ooyala.

Netflix, Amazon, and YouTube have all been tweaking their recommendation engines for years. Ingesting user data and predicting what each consumer is most likely to want to watch sounds relatively simple -- but getting it right is a different story.

advertisement

advertisement

Moving further into AI capabilities, marketers can learn to connect disparate video ads based on themes, personalities and tone. This is where AI can really impress.

Microsoft Azures cloud computing product, which has been integrated into Ooyalas Flex platform, takes video content and boils it down to natural language -- covering people, colors, actions, logos, even the type of event on screen -- using the data to inform what paid media would be appropriate to serve against the content.

With near-immediate access to words spoken in any broadcast setting, marketers and advertisers can develop a deep understanding of themes present in a newscast, TV show or other video event, explained Martin Wahl, the principal product manager for Microsoft Azure. For example, Sinclair pays employees up to $85 an hour to transcribe all shows on its properties, said Wahl. AI can do the same job [and] link related words, making everything immediately searchable for a tenth or hundredth of the price it costs to have a human do it.

Even translations can be done in next to real time, opening up a completely new opportunity for international distribution of live content. Importantly, it can open up opportunities for national advertisers to expand into the international space more seamlessly.

These capabilities will save both time and money, providing marketers with immediate insights on what kinds of ad creatives are most appropriate to serve after a particular segment. With robust metadata, the AI can even suggest which ads to buy.

As Wahl put it, these capabilities are a strong start, but the real difference-maker will be how marketers decide to use the collected insights in novel ways that are yet to be discovered. That is when we will begin to see the true value of AI in marketing. Beyond artificial intelligence, human intelligence will continue to play a central role in harnessing AI.

Link:

AI: The Future Of Digital Marketing (And Everything Else) - MediaPost Communications

Once upon a time, terrorists used bombs, machetes, and bullets to get their message across. While that's still the case, modern day terror has a new tool at its disposal, one that it has become particularly adept and successful at deploying: social media. This stark reality has come to light in the wake of terror campaigns that ended with participants pledging their support to their chosen causes and posting them on social media platforms.

Other insidious forms of communication and objectionable material have flourished in the internet era as well. Hate speech and violent threats have found homes there. Governments and advertisers worldwide are aware of the phenomenon and are increasingly pressuring social-media companies like Facebook, Inc. (NASDAQ:FB), Alphabet Inc. (NASDAQ:GOOGL) (NASDAQ:GOOG), Twitter, Inc. (NYSE:TWTR), and Microsoft Corporation (NASDAQ:MSFT) to police undesirable content on their sites.

The sheer volume of content and the differences and complexity of local laws and regulations conspired to create a near-insurmountable task for these sites. However, recent advances in artificial intelligence (AI) are being brought to bear, and producing surprisingly effective results.

Facebook is deploying AI to fight terror. Image source: Facebook.

Facebook revealed that new AI algorithms based on image recognition have been deployed to assist with the Herculean chore. One tool has been developed to scan the site for images and live videos containing terrorist propaganda, including beheadings, and to remove them without the intercession of a human moderator.

Another system has been trained to identify accounts that have been set up by terrorists, and prevent them from setting up additional accounts. Another algorithm is being trained in the language of propaganda to help identify posts related to terror. Once the content has been identified and removed, the system catalogs the data, then consistently scans the site and identifies attempts to repost it.

Twitter has been deploying similar tools based on AI for rooting out terrorist content. The company says that these methods flagged 74% of the nearly 377,000 accounts it removed between July and December of 2016.

This follows an alliance by some of the biggest names in tech circles late last year to create a database of the worst content, to prevent it from being reposted on any of the sites. YouTube, Twitter, and Facebook joined Microsoft in the venture to create unique digital identifiers, or "fingerprints," to use for automatically detecting and removing content that had previously been tagged as terrorist propaganda.

Microsoft developed and deployed similar technology to battle child pornography on the internet. The system was used to detect, report, and remove the images contained in a database.

Big tech is bringing AI to the fight on terror. Image source: Getty Images.

Google, the Alphabet subsidiary and owner of YouTube, is a pioneer in AI and recently found another way to use the nascent technology. YouTube faced a massive boycott from some of its biggest advertisers after it was revealed that brand advertising had appeared on YouTube videos containing racist, homophobic, anti-Semitic, and terrorist content. The company applied new AI techniques to the task, and within weeks achieved a 500% improvement in identifying objectionable content. YouTube revealed that more than half the content it removed over the previous six months for containing terrorist-related material had been identified using AI.

The world is a complicated place, and new technology brings new challenges. The advent of social media brought the world closer together, for better or for worse. Artificial intelligence is still a nascent technology, and while it isn't a panacea, it is being used in a variety of ways that make the world a better place.

Suzanne Frey, an executive at Alphabet, is a member of The Motley Fool's board of directors. Teresa Kersten is an employee of LinkedIn and is a member of The Motley Fool's board of directors; LinkedIn is owned by Microsoft. Danny Vena owns shares of Alphabet (A shares) and Facebook. Danny Vena has the following options: long January 2018 $640 calls on Alphabet (C shares) and short January 2018 $650 calls on Alphabet (C shares). The Motley Fool owns shares of and recommends Alphabet (A shares), Alphabet (C shares), Facebook, and Twitter. The Motley Fool has a disclosure policy.

Read the rest here:

How Social Media Is Using AI to Fight Terrorism - Motley Fool

You see, Mattheij decided he wanted in on the profitable cottage industry of online Lego reselling, and after placing a bunch of bids for the colorful little blocks on eBay, he came into possession of 2 tons (4,400 pounds) of Lego -- enough to fill his entire garage.

As Mattheij explains in his blog post, resellers can make up to 40 ($45) per kilogram for Lego sets, and rare parts and Lego Technic can fetch up to 100 ($112) per kg. If you really want to rake in the cash, however, you have to go through the exhaustive process of manually sorting through your bulk Lego before selling it in smaller groupings online. Instead of spending an eternity sifting through his own, intimidatingly large collection, Mattheij set to work on building an automated Lego sorter powered by a neural network that could classify the little building blocks. In case you were wondering, Lego comes in more than 38,000 shapes and over 100 shades of color, which amounts to a lot of sorting even with the aid of AI.

Starting with a proof of concept (built using Lego, naturally), Mattheij spent the following six months improving upon his prototype with a lot of DIY handiwork. In his own words, he describes his present setup as a "hodge-podge of re-purposed industrial gear" stuck together using "copious quantities of crazy glue" and a "heavily modified" home treadmill.

The current incarnation uses conveyor belts to carry the Lego past a web camera that is set up to take images of the blocks. These are then fed to the neural network as part of its classification training, and all Mattheij has to do is spot the errors in its judgement.

"As the neural net learns, there are fewer mistakes, and the labeling workload decreases," he states. "By the end of two weeks I had a training data set of 20,000 correctly labeled images."

With his prototype up and running, Mattheij claims he is just waiting for the machine learning software to reliably class all of the images itself, and then he can start selling off the lucrative toy. If Matthiej manages to get the system working, he could then rechannel those profits into new expensive Lego projects.

Read the original here:

Sorting Lego sucks, so here's an AI that does it for you - Engadget

If we ever want future robots to do our bidding, theyll have to understand the world around them in a complete wayif a robot hears a barking noise, whats making it? What does a dog look like, and what do dogs need?

AI research has typically treated the ability to recognize images, identify noises, and understand text as three different problems, and built algorithms suited to each individual task. Imagine if you could only use one sense at a time, and couldnt match anything you heard to anything you saw. Thats AI today, and part of the reason why were so far from creating an algorithm that can learn like a human. But two new papers from MIT and Google explain first steps for making AI see, hear, and read in a holistic wayan approach that could upend how we teach our machines about the world.

It doesnt matter if you see a car or hear an engine, you instantly recognize the same concept. The information in our brain is aligned naturally, says Yusuf Aytar, a post-doctoral AI research at MIT who co-authored the paper.

That word Aytar usesalignedis the key idea here. Researchers arent teaching the algorithms anything new, but instead creating a way for them to link, or align, knowledge from one sense to another. Aytar offers the example of a self-driving car hearing an ambulance before it sees it. The knowledge of what an ambulance sounds like, looks like, and its function could allow the self-driving car to prepare for other cars around it to slow down, and move out of the way.

To train this system, the MIT group first showed the neural network video frames that were associated with audio. After the network found the objects in the video and the sounds in the audio, it tried to predict which objects correlated to which sounds. At what point, for instance, do waves make a sound?

Next, the team fed images with captions showing similar situations into the same algorithm, so it could associate words with the objects and actions pictured. Same idea: first the network separately identified all the objects it could find in the pictures, and the relevant words, and then matched them.

The network might not seem incredibly impressive from that descriptionafter all, we have AI that can do those things separately. But when trained on audio/images and images/text, the system was then able to match audio to text, when it had never been trained to know which words correspond to different sounds. Researchers claim this indicated the network had built a more objective idea of what it was seeing, hearing, or reading, one that didnt entirely rely on the medium it used to learn the information.

One algorithm that can align its idea of an object across sight, sound, and text can automatically transfer what its learned from what it hears to what it sees. Aytar offers the examples that if the algorithm hears a zebra braying, it assumes that a zebra is similar to a horse.

It knows that [the zebra] is an animal, it knows that it generates these kinds of sounds, and kind of inherently it transfers this information across modalities, Aytar says. These kinds of assumptions allow the algorithm to make new connections between ideas, strengthening its understanding of the world.

Googles model behaves similarly, except with the addition of being able to translate text as well. Google declined to provide a researcher to talk more about how its network operated. However, the algorithm has been made available online to other researchers.

Neither of these techniques from Google or MIT actually performed better than the single-use algorithms, but Aytar says that this wont be the case for long.

If you have more senses, you have more accuracy, he said.

Originally posted here:

MIT and Google researchers have made AI that can link sound, sight, and text to understand the world - Quartz

Andrew Ng, the former chief scientist of Baidu, announced his next venture, Deeplearning.ai, with only a logo, a domain name and a footnote pointing to an August launch date. In an interesting twist, the Deeplearning.ai domain name appears to be registered to Baidus Sunnyvale AI research campus the same office Ng would have worked out of as an employee.

Its unclear whether Ng began his work on Deeplearning.ai while still an employee at Baidu. According to data pulled from theWayback Machine, the domain was parked at Instra and picked up sometime between 2015 and 2017.

Registering that domain to Baidu accidentally would be an amateur mistake and registering it intentionally just leaves me with more unanswered questions. Im left wondering about the relationship between Baidu and Deeplearning.ai and its connection to Andrew Ngs departure. Of course, its also possible that there was some sort of error that caused an untimely mistake.

UPDATE: Baidu provided us the following response.

Baidu has no association with this project but we wish Andrew the best in his work.

Ng left the company in late March of this year, promising to continue his work of bringing the benefits of AI to everyone. Baidu is known for having unique technical expertise in natural language processing and its recently been putting resources into self-driving cars and other specific deep learning applications.

It makes sense that Ng would take advantage of his name recognition to raise a large round to maximize his impact on the machine intelligence ecosystem. I cant see a general name like Deeplearning.ai being used to sell a self-driving car company or a verticalized enterprise tool. Its more likely that Ng is building an enabling technology that aims to become critical infrastructure to support the adoption of AI technologies.

While this could technically encompass specialized hardware chips for deep learning, Im more inclined to bet that it is a software solution given Ngs expertise. Google CEO Sundar Pichai made a splash back at I/O last month when he discussed AutoML the companys research work to automate the design process of neural networks. If I was going to come up with a name for a company that would build on, and ultimately commercialize, this technology, it would be Deeplearning.ai.

This is super speculative, but I think it might be an AI tool to help generate AI training data sets or something else that will accelerate the development of AI models and products, Malika Cantor, partner at AI investment firm Comet Labs told me. Im very excited about having more tools and platforms to support the AI ecosystem.

Prior to his time at Baidu, Ng was instrumental in building out the Google Brain Team, one of the companys core AI research groups.Ng is a highly respected researcher and evangelist in the AI space with connections spanning industries and geographic borders. If Ng truly believes that AI is the new electricity, he will surely try to position Deeplearning.ai to take advantage of the windfall.

Weve reached out to both Baidu and Andrew Ng and will update this post if we receive additional information.

Originally posted here:

Andrew Ng announces Deeplearning.ai, his new venture after ... - TechCrunch

While the growth of deep neural networks has helped propel the field of machine learning to new heights, theres still a long road ahead when it comes to creating artificial intelligence. Thats the message from a panel of leading machine learning and AI experts who spoke at the Association for Computing Machinerys Turing Award Celebration conference in San Francisco today.

Were still a long way off from human-level AI, according to Michael I. Jordan, a professor of computer science at the University of California, Berkeley. He said that applications using neural nets are essentially faking true intelligence but that their current state allows for interesting development.

Some of these domains where were faking intelligence with neural nets, were faking it well enough that you can build a company around it, Jordan said. So thats interesting, but somehow not intellectually satisfying.

Those comments come at a time of increased hype for deep learning and artificial intelligence in general, driven by interest from major technology companies like Google, Facebook, Microsoft, and Amazon.

Fei-Fei Li, who works as the chief scientist for Google Cloud, said that she sees this as the end of the beginning for AI, but says there are still plenty of hurdles ahead. She identified several key areas where current systems fall short, including a lack of contextual reasoning, a lack of contextual awareness of their environment, and a lack of integrated understanding and learning.

This kind of euphoria of AI has taken over, and [the idea that] weve solved most of the problem is not true, she said.

One pressing issue identified by Raquel Urtasun, who leads Ubers self-driving car efforts in Canada, is that the algorithms used today dont model uncertainty very well, which can prove problematic.

So they will tell you that there is a car there, for example, with 99 percent probability, and they will tell you the same thing whether they are wrong or not, she said. And most of the time they are right, but when they are wrong, this is a real issue for things like self-driving [cars].

The panelists did concur that an artificial intelligence that could match a human is possible, however.

I think we have at least half a dozen major breakthroughs to go before we get close to human-level AI, said Stuart Russell, a professor of computer science and engineering at the University of California, Berkeley. But there are very many very brilliant people working on it, and I am pretty sure that those breakthroughs are going to happen.

Continue reading here:

AI is still several breakthroughs away from reality - VentureBeat