When manufacturers strategically deploy artificial intelligence (AI) into their operating environments, the results can be game changing. Setting the stage for constant innovation. Avoiding down production lines. Anticipating and taking action on dramatic market changes before the competition. The reality for most manufacturers? AI deployments can be challenging. Getting it right takes determining and dedication to optimizing the process.
I recently connected with Landing AI's Quinn Killough to discuss what happens, and how manufacturers can move past the obstacles.
IW: When it comes to adopting AI, where do most manufacturers run into problems?
Killough: One of the major points of failure in AI projects for manufacturers is the jump from a working model in the lab to a working model in production. Many manufacturers have high hopes for AI improving automation on their plant floors, and once theyve honed-in on a valuable problem/project, they go to work in the lab and start training AI models. After just a few iterations (depending on the complexity of the problem) they may find their model is performing at the accuracy level desired.
They think, great! AI is magic and its going to solve all of our problems, and decide to take the solution to production so they can start reaping the benefits. The issue that we have seen with countless manufacturers across a number of industries is that the models performance in production is worse than it was in the lab and is not yet suitable for true deployment.
They end up needing to go back to the lab and the drawing board to iterate on the model further and in the meantime revert back to their old way of doing things on the floor. As we see it, this jump from a PoC in the lab to a fully functional, value adding, deployed model is much harder than expected. Not only is this an issue for the project, i.e. longer development times, more money spent, opportunity cost of the old system running longer, but it is also an issue of perception for AI - people can lose faith after seeing this process unfold.Landing AI's Quinn Killough
IW: Why do manufacturers struggle to get past that point?
Killough: Frequently, the number one contributor to this struggle is data quality. A machine learning model is only as good as the data you are putting into it. Machine learning engineers spend up to 80% of their time prepping their data, and there is a lot that can go wrong in this phase. Issues with data quality can come in many flavors, but the two we see the most often are poorly labeled data, and bad dataset distribution.
A model will struggle in production if the manufacturer fails to accurately label the data. When you dont have millions of datasets to wash out the impact of labeling mistakes, just one or two improperly labeled items can lead to a model that does not perform well. In the case of visual defect detection in manufacturing, a common contributor to bad labels is inadequate defect definitions. It is not uncommon for two subject matter experts on a plant floor to disagree on whether a defect is a scratch or a crack, which in turn makes it difficult for the people doing labeling to label images correctly.
When machine learning engineers are collecting data to train their model, the data collected often is not diverse enough to represent the variety of edge cases that would actually be seen in the production environment. Because of this, edge cases that are not seen frequently may be underrepresented when training and testing so the model ends up not being great at finding these. Lets consider a visual inspection project where a manufacturer is looking for several categories of defects on a phone screen - dust, dead pixels and cracks. Dust and dead pixels are super common, so they have a ton of data on these categories. Cracks, on the other hand, are a very infrequent defect and therefore there is less data and the model is not great at recognizing them. A big issue here is that oftentimes the infrequent defects are some of the most critical ones to catch. If this manufacturer were to test on the same distribution as they trained they would get good results. But when going to production they would quickly find out they cant perform on their most critical defects and be forced to pull the system.
IW: What are the keys to getting past the barriers?
Killough: There are a couple of things that can be done to help get past these barriers and achieve a successful deployed model.
First, create an airtight labeling process, and second, move the project out of the lab sooner. This starts with data collection and making sure your dataset is as representative of the production process as possible. After a well-balanced dataset is established, a systematic method for driving agreement on what the definitions of the label categories are needs to be incorporated. Consensus tasks have proven to be a great way to achieve this - i.e. create a system where multiple experts review that same data and compare labels, and after this, discrepancies can be resolved in order to make label definitions less ambiguous for labelers. Now that labelers have the clearest instructions possible, they have a good chance at labeling the data accurately, but they are still human and will likely make mistakes - this is why we would recommend having a tight review process. This means that every single piece of labeled data is reviewed. It may seem like overkill, but a single mistake can be costly on model performance. Establishing this airtight data preparation process will inherently improve model performance and help project leaders avoid costly pitfalls and mistakes.
Outside of having an airtight data preparation process, we also recommend utilizing approaches that allow you to get from the lab bench to the production floor faster than most would think is intuitive. Often teams want to achieve their ultimate goal for model performance - lets say 95% accuracy is the goal - in the lab before moving out to production, but this can be flawed. There will inevitably be unforeseen changes as well as the edge cases mentioned earlier that pop up when deployed to production that hurts performance. Instead, if we spend less time in the lab to get to a lower bar of 80%, we can then deploy that model in production and start uncovering those production edge cases and issues much sooner. To do this, AI teams can use two approaches - shadow mode or putting a human in the loop. Shadow mode means running a model on actual production data without the predictions of the models impacting production - meaning your existing process is not impacted while at the same time you can iterate and improve your model.
Putting a human in the loop is another option if youd still like to benefit from an underperforming model while it is being improved. In this situation the model would run as though it was actually deployed, and a human would review each low confidence prediction. Now a low performing model can be deployed earlier, 100% of parts can still be inspected, and the model can be iterated upon to continuously do more and more of the work until fully autonomous.
IW: Once manufacturers are able to adopt AI, what steps can they take to optimize its use?
Killough: Continuous learning is the most important aspect to optimizing a deployed solution. This entails a few different aspects including data collection, inference monitoring, and a method for retraining.
Model deployment is just the beginning. It takes a solid system just to maintain current performance of a model, let alone further performance optimization. The first step here is having a method for continuous data collection. As the model is making predictions its important to be able to collect data where it may have struggled. With an inference monitoring system where operators can view real time and past predictions this process can be fairly simple.
While collecting and labeling data, a periodic retraining of the model is also very important. This periodic retraining on new data helps to optimize performance by catching and training on those infrequent edge cases. Also, by continuously retraining on the latest state of the production system you can account for naturally occurring changes in production. For example, in optical inspections, if a material changes in your production it could impact model performance due to a new appearance of the product, but not enough to set off any alarms or flag anyones attention. Having a quick and easy way to retrain and roll out a new model is important when trying to optimize or maintain an AI system - without the help of automation this task can be rather time consuming.
Original post:
AI Adoption Hitting Barriers? Here's How to Get to the Next Level - IndustryWeek
- Classic reasoning systems like Loom and PowerLoom vs. more modern systems based on probalistic networks [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Using Amazon's cloud service for computationally expensive calculations [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Software environments for working on AI projects [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- New version of my NLP toolkit [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Semantic Web: through the back door with HTML and CSS [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Java FastTag part of speech tagger is now released under the LGPL [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Defining AI and Knowledge Engineering [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Great Overview of Knowledge Representation [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Something like Google page rank for semantic web URIs [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- My experiences writing AI software for vehicle control in games and virtual reality systems [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- The URL for this blog has changed [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- I have a new page on Knowledge Management [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- N-GRAM analysis using Ruby [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Good video: Knowledge Representation and the Semantic Web [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Using the PowerLoom reasoning system with JRuby [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Machines Like Us [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- RapidMiner machine learning, data mining, and visualization tool [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- texai.org [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- NLTK: The Natural Language Toolkit [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- My OpenCalais Ruby client library [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Ruby API for accessing Freebase/Metaweb structured data [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Protégé OWL Ontology Editor [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- New version of Numenta software is available [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Very nice: Elsevier IJCAI AI Journal articles now available for free as PDFs [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Verison 2.0 of OpenCyc is available [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- What’s Your Biggest Question about Artificial Intelligence? [Article] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Minimax Search [Knowledge] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Decision Tree [Knowledge] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- More AI Content & Format Preference Poll [Article] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- New Planners Solve Rescue Missions [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Neural Network Learns to Bluff at Poker [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Pushing the Limits of Game AI Technology [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Mining Data for the Netflix Prize [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Interview with Peter Denning on the Principles of Computing [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Decision Making for Medical Support [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Neural Network Creates Music CD [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- jKilavuz - a guide in the polygon soup [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Artificial General Intelligence: Now Is the Time [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Apply AI 2007 Roundtable Report [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- What Would You do With 80 Cores? [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Software Finds Learning Language Child's Play [News] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Artificial Intelligence in Games [Article] [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Artificial Intelligence Resources [Last Updated On: November 8th, 2009] [Originally Added On: November 8th, 2009]
- Alan Turing: Mathematical Biologist? [Last Updated On: April 25th, 2012] [Originally Added On: April 25th, 2012]
- BBC Horizon: The Hunt for AI ( Artificial Intelligence ) - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Can computers have true artificial intelligence" Masonic handshake" 3rd-April-2012 - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Kevin B. Korb - Interview - Artificial Intelligence and the Singularity p3 - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Artificial Intelligence - 6 Month Anniversary - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Science Breakthroughs [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Hitman: Blood Money - Part 49 - Stupid Artificial Intelligence! - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Research Members Turned Off By HAARP Artificial Intelligence - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Artificial Intelligence Lecture No. 5 - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- The Artificial Intelligence Laboratory, 2012 - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Charlie Rose - Artificial Intelligence - Video [Last Updated On: April 30th, 2012] [Originally Added On: April 30th, 2012]
- Expert on artificial intelligence to speak at EPIIC Nights dinner [Last Updated On: May 4th, 2012] [Originally Added On: May 4th, 2012]
- Filipino software engineers complete and best thousands on Stanford’s Artificial Intelligence Course [Last Updated On: May 4th, 2012] [Originally Added On: May 4th, 2012]
- Vodafone xone™ Hackathon Challenges Developers and Entrepreneurs to Build a New Generation of Artificial Intelligence ... [Last Updated On: May 4th, 2012] [Originally Added On: May 4th, 2012]
- Rocket Fuel Packages Up CPG Booster [Last Updated On: May 4th, 2012] [Originally Added On: May 4th, 2012]
- 2 Filipinos finishes among top in Stanford’s Artificial Intelligence course [Last Updated On: May 5th, 2012] [Originally Added On: May 5th, 2012]
- Why Your Brain Isn't A Computer [Last Updated On: May 5th, 2012] [Originally Added On: May 5th, 2012]
- 2 Pinoy software engineers complete Stanford's AI course [Last Updated On: May 7th, 2012] [Originally Added On: May 7th, 2012]
- Percipio Media, LLC Proudly Accepts Partnership With MIT's Prestigious Computer Science And Artificial Intelligence ... [Last Updated On: May 10th, 2012] [Originally Added On: May 10th, 2012]
- Google Driverless Car Ok'd by Nevada [Last Updated On: May 10th, 2012] [Originally Added On: May 10th, 2012]
- Moving Beyond the Marketing Funnel: Rocket Fuel and Forrester Research Announce Free Webinar [Last Updated On: May 10th, 2012] [Originally Added On: May 10th, 2012]
- Rocket Fuel Wins 2012 San Francisco Business Times Tech & Innovation Award [Last Updated On: May 13th, 2012] [Originally Added On: May 13th, 2012]
- Internet Week 2012: Rocket Fuel to Speak at OMMA RTB [Last Updated On: May 16th, 2012] [Originally Added On: May 16th, 2012]
- How to Get the Most Out of Your Facebook Ads -- Rocket Fuel's VP of Products, Eshwar Belani, to Lead MarketingProfs ... [Last Updated On: May 16th, 2012] [Originally Added On: May 16th, 2012]
- The Digital Disruptor To Banking Has Just Gone International [Last Updated On: May 16th, 2012] [Originally Added On: May 16th, 2012]
- Moving Beyond the Marketing Funnel: Rocket Fuel Announce Free Webinar Featuring an Independent Research Firm [Last Updated On: May 23rd, 2012] [Originally Added On: May 23rd, 2012]
- MASA Showcases Latest Version of MASA SWORD for Homeland Security Markets [Last Updated On: May 23rd, 2012] [Originally Added On: May 23rd, 2012]
- Bluesky Launches Drones for Aerial Surveying [Last Updated On: May 23rd, 2012] [Originally Added On: May 23rd, 2012]
- Artificial Intelligence: What happened to the hunt for thinking machines? [Last Updated On: May 25th, 2012] [Originally Added On: May 25th, 2012]
- Bubble Robots Move Using Lasers [VIDEO] [Last Updated On: May 25th, 2012] [Originally Added On: May 25th, 2012]
- UHV assistant professors receive $10,000 summer research grants [Last Updated On: May 27th, 2012] [Originally Added On: May 27th, 2012]
- Artificial intelligence: science fiction or simply science? [Last Updated On: May 28th, 2012] [Originally Added On: May 28th, 2012]
- Exetel taps artificial intelligence [Last Updated On: May 29th, 2012] [Originally Added On: May 29th, 2012]
- Software offers brain on the rain [Last Updated On: May 29th, 2012] [Originally Added On: May 29th, 2012]
- New Dean of Science has high hopes for his faculty [Last Updated On: May 30th, 2012] [Originally Added On: May 30th, 2012]
- Cognitive Code Announces "Silvia For Android" App [Last Updated On: May 31st, 2012] [Originally Added On: May 31st, 2012]
- A Rat is Smarter Than Google [Last Updated On: June 5th, 2012] [Originally Added On: June 5th, 2012]