Scientists develop new machine learning method that can make robots safer. New method provides simpler and more intuitive models of physical situations.
Understanding how a robot will react under different conditions is essential to guaranteeing its safe operation. But how do you know what will break a robot without actually damaging it? A new method developed by scientists at the Institute of Science and Technology Austria (IST Austria) and the Max Planck Institute for Intelligent Systems (MPI for Intelligent Systems) is the first machine learning method that can use observations made under safe conditions to make accurate predictions for all possible conditions governed by the same physical dynamics. Especially designed for real-life situations, their method provides simple, interpretable descriptions of the underlying physics. The researchers will present their findings tomorrow at this year’s prestigious International Conference for Machine Learning (ICML).
In the past, machine learning was only capable of interpolating data—making predictions about situations that are “between” other, known situations. It was incapable of extrapolating—making predictions about situations outside of the known—because it learns to fit the known data as closely as possible locally, regardless of how it performs outside of these situations. In addition, collecting sufficient data for effective interpolation is both time- and resource-intensive, and requires data from extreme or dangerous situations. But now, Georg Martius, former ISTFELLOW and IST Austria postdoc, and since 2017 a group leader at MPI for Intelligent Systems in Tübingen, Subham S. Sahoo, a PhD student also at MPI for Intelligent Systems, and Christoph Lampert, professor at IST Austria, developed a new machine learning method that addresses these problems, and is the first machine learning method to accurately extrapolate to unseen situations.
The key feature of the new method is that it strives to reveal the true dynamics of the situation: it takes in data and returns the equations that describe the underlying physics. “If you know those equations,” says Georg Martius, “then you can say what will happen in all situations, even if you haven’t seen them.” In other words, this is what allows the method to extrapolate reliably, making it unique among machine learning methods.
The team’s method sets itself apart in several other ways as well. First, the final approximations previously produced during machine learning were far too complex for a human to understand or work with. In the new method, the resulting equations are far simpler: “Our method’s equations are something you would see in a textbook—simple and intuitive,” says Christoph Lampert. The latter is another key difference: other machine learning methods give no insight into the relationship between conditions and results—and thus, no intuition on whether the model is even plausible. “In every other area of research, we expect models that make physical sense, that tell us why,” adds Lampert. “This is what we should expect from machine learning, and what our method provides.” Finally, in order to guarantee interpretability and optimize for physical situations, the team based their learning method on a different type of framework. This new design is simpler than previous methods, which in practice means that less data is needed to give the same or even better results.
And it’s not all theory: “In my group, we’re actually working on developing a robot that uses this type of learning. In the future, the robot would experiment with different motions, then be able to use machine learning to uncover the equations that govern its body and movement, allowing it to avoid dangerous actions or situations,” adds Martius. While robots are one active area of research, the method can be used with any type of data, from biological systems to X-ray transition energies, and can also be incorporated into larger machine learning networks.
Receive an email update when we add a new MACHINE LEARNING article.
The Latest on: Machine learning
via Google News
The Latest on: Machine learning
- Brickschain Launches AI and Machine Learning Platform Synapse at Autodesk University on November 12, 2018 at 4:52 pm
LAS VEGAS, November 12, 2018 (Newswire.com) - Brickschain announces the release of Synapse, the long-anticipated artificial intelligence platform for the built environment. It gives construction compa... […]
- Secret Escapes: How machine learning reduced cost per lead by 38% [case study] on November 12, 2018 at 4:50 pm
The digital travel sector continues to grow, with increased competition across channels resulting in more expensive CPL (cost per lead) and CPA (cost per acquisition). So, when it comes to advertising ... […]
- Slidecast: BigDL Open Source Machine Learning Framework for Apache Spark on November 12, 2018 at 3:59 pm
In this video, Beenish Zia from Intel presents: BigDL Open Source Machine Learning Framework for Apache Spark. BigDL is a distributed deep learning library for Apache Spark*. Using BigDL, you can writ... […]
- Machine learning for pharma marketers: The time is now to budget, experiment and adjust, expert says on November 12, 2018 at 2:54 pm
When does CMI/Compas executive Paul Kallukaran think pharma companies should start looking at using machine learning in their marketing strategies? "Yesterday," he said at a recent digital conference. ... […]
- How Machine Learning Could Help California Fight Wildfires on November 12, 2018 at 12:24 pm
Wildfires have torn through nearly hundreds of thousands of acres in California already this year, forcing evacuations and causing dozens of deaths. A new story by the Bay Area News Group looks ... […]
- How Facebook Flags Terrorist Content With Machine Learning on November 12, 2018 at 10:54 am
For years, content that promotes terrorism has thrived on social media platforms like Facebook and Twitter. Facebook engineered an algorithm that assigns each post a score based on the ... […]
- Top 10 Machine Learning, Deep Learning, and Data Science Courses for Beginners (Python and R) on November 12, 2018 at 10:32 am
Did you know that 50- 80% of your enterprise business processes can be automated with AssistEdge? Identify processes, deploy bots and scale effortlessly with AssistEdge. Data Science, Machine ... […]
- VoltDB and MapR Technologies Join Forces to Support Machine Learning for Real-Time Decision Making on November 12, 2018 at 8:40 am
BEDFORD, Mass., Nov. 12, 2018 /PRNewswire/ -- VoltDB, the enterprise-class database that powers real-time intelligent decisions on streaming data, today announced a collaboration with MapR Technologie... […]
- How Facebook uses machine learning to fight ISIS and Al-Qaeda propaganda on November 12, 2018 at 6:17 am
Facebook claims to have proactively found and removed 99 percent of terrorist-related content on the site for the past three quarters. It’s given some insight into its processes in a blog post. Some s... […]
via Bing News