CAREER

Zamani CAREER award to bridge the gap between industry and academia in autonomous systems

Anonymous — Fri, 24 Jun 2022 18:53:20 +0000

Zamani CAREER award to bridge the gap between industry and academia in autonomous systems Anonymous (not verified) Fri, 06/24/2022 - 12:53

Grace Wilson

Majid Zamani, an assistant professor in the Department of Computer Science at CU �鶹ӰԺ, wants to use real-life data, rather than mathematical models, to study and control autonomous systems with both software and physical components, bridging the gap between academia and industry and ensuring safety for all users.

He has just been presented with a prestigious CAREER award from the National Science Foundation (NSF) for his proposal entitled “A Data-Driven Approach for Verification and Control of Cyber-Physical Systems.”

CAREER awards provide funding over five years to support the research and educational activities of early career faculty members who have the potential to become leaders in their field. Six faculty members within the College of Engineering and Applied Science received CAREER Awards from the National Science Foundation in 2022.

Zamani said his CAREER award unifies three different fields: formal methods in computer science, optimization in operation research and control theory. The research brings insight from each to understand how to verify the safety of autonomous systems purely through data analysis.

"If I have enough data collected, I can work directly with the data to systematically generate the software code in charge of controlling a system," Zamani said.

Currently, rigid mathematical models that describe the behaviors of a system are the main ingredients of most academic research in ensuring safety in cyber-physical systems–- systems where software interacts tightly with physical systems—such as self-driving cars, pacemakers and power networks.

To have these mathematical models, someone must rigorously model every part of the system. When you have thousands of different components in a machine and possibly hundreds of computer program interactions, the layers of complexity stack exponentially and it is very hard to build the models accurately. Even if the models are computed, Zamani said, they are too complex to be dealt with.

With his CAREER award, Zamani will be working to entirely bypass the need for such a model of the system. This means that systems that are too complex for us to know their internal workings, known as "black boxes," can still be formally guaranteed as safe.

Safety is a constant concern when computer programs can impact the physical world. A single catastrophic safety failure in a cyber-physical system could cause trust in the autonomous system to be lost and lead to loss of life or infrastructure.

Despite the need for safety, many self-driving car industries do not have the time or interest to mathematically model their systems, Zamani said.

"You approach a company and they say, 'no, we don't have a model. We have the actual car or its simulator, but we don't know the precise mathematical model for it,'" Zamani said.

Zamani's award centers around the recent advances in inexpensive sensor technologies that can gather large amounts of data from a system's behaviors as it is run without autonomy, like when a person drives a car destined for autopilot.

Zamani said that, while they may not have models, industry partners do have large amounts of data available, making it possible to rigorously analyze realistic systems and build a "controller," the software code that autonomously controls the system, such as an auto-pilot in a self-driving car.

The framework that Zamani is crafting is also not system-dependent. Rather than needing a separate way of understanding self-driving cars, drones or medical devices, his work is abstracting the logic needed to create algorithms for controlling all these types of systems.

In addition, safety can mean different things to different people. Zamani's work allows companies to decide how conservative their safety confidence levels should be. The more data collected, the higher the confidence levels Zamani's framework is able to guarantee.

And, as well as determining what level of safety is necessary, the research supports a variety of "properties of interest." For example, if a car is safe only if it doesn't crash, it might speed past the speed-limit regularly, but by adding a property that requires the car to also follow the speed-limit, you craft a controller that accommodates both properties of interest.

This system-agnostic, flexible and data-driven framework provides an alternative to the severe computational complexity of rigid mathematical models and strong assumptions made about them that have caused a divide between academia and industry.

"The main goal of my CAREER award is closing the gap between what happens in reality and the theoretical, rigorous analyses which happen in academia. People in industry are not using the techniques we've been developing in academia. There is a gap between their assumptions and ours, and this work is trying to help close it." Zamani said.

Zamani wants to use real-life data, rather than mathematical models, to study and control autonomous systems with both software and physical components, bridging the gap between academia and industry and ensuring safety for all users.

Traditional

White

Trivedi seeks to democratize artificial intelligence through CAREER award

Anonymous — Thu, 23 Jun 2022 20:05:18 +0000

Trivedi seeks to democratize artificial intelligence through CAREER award Anonymous (not verified) Thu, 06/23/2022 - 14:05

Grace Wilson

Ashutosh Trivedi, an assistant professor in the Department of Computer Science at CU �鶹ӰԺ, is working to democratize artificial intelligence by making machine learning more programmable, trustworthy and accessible to everyone.

He has just been presented with a prestigious CAREER award from the National Science Foundation to do so. The award supports the research and educational activities of early career faculty members who have the potential to become leaders in their field. Trivedi's provides $600,000 over the next five years. Six faculty members within the College of Engineering and Applied Science received CAREER Awards from the National Science Foundation in 2022.

Trivedi will use the award to improve abstraction as an alternative to traditional neural networks, which have huge energy and data requirements, and to build our ability to trust and understand machine learning.

Trivedi, who was born in India and raised during the country's intensive focus on computers in the 80s and 90s, knows how essential access is. Reminiscing about this pivotal moment in India's history, Trivedi said, "by having the power to talk to computers, we transformed not only our own lives, but those around us."

If people everywhere aren't given access to artificial intelligence now, he said, it will remain confined to applications with high capital investment, rather than being a vehicle for widespread innovative problem-solving.

"Computers can be little engines of creativity and they can co-create with us. Humans are not the only sources of beauty and ingenuity," Trivedi said.

To understand the fundamental change Trivedi is pursuing in his research, we must first understand what machine learning looks like right now. This current moment, he said, is as transformational as the shift from computers that took up several rooms to those sitting in a wrist-watch today.

Right now, many applications rely on neural networks and reinforcement learning.

Neural networks are large computer programs that, given huge amounts of data, create working definitions for what they have been trained on. For example, after viewing a large number of cat images, the machine "learns" to see a grouping of pixels as a cat.

But, Trivedi said, there's a problem with that. The program can't explain what a cat is beyond that grouping of pixels.

“If you train something with a neural network, you do not know what has been learned. We cannot explain why something is a cat or not a cat," he said. The machine's learning process and reasoning is hidden from us.

The high skill-floor and intense resource demand of these large neural networks can also keep machine learning away from passionate but under-resourced programmers who can’t afford the massive costs of creating and maintaining the networks.

Reinforcement learning, especially when combined with neural networks, is a promising machine learning approach to problem solving if it could be made more trustworthy and capable of solving complex problems.

Reinforcement learning is similar to training a dog, Trivedi said. By rewarding good behaviors over and over and chastising the bad ones, you can slowly train a dog to do many tricks, like shake hands or heel.

But when and how should a reward be given? In computer science, this is a question programmers must answer each time they create a reinforcement learning application. If you reward at the wrong time, you could cause a program to learn the wrong thing, like mis-training a dog to bark when it sees food.

A programmer could create a program that inadvertently damages a power grid or makes racially-biased decisions on who can access a home loan due to a bad internal reward system.

Trivedi's CAREER proposal focuses on building tools for reinforcement learning that free the programmer from the burden of translating desired outcomes to specific rewards. Instead of using a gut-feeling for the reward, programmers would now have a rigorous, formal system to assist them that they can trust.

But, even if rewards are given correctly, reinforcement learning traditionally doesn't work as well for the large, complicated problems machine learning has so much promise for.

So Trivedi wants to be able to increase the scale of tasks that reinforcement learning can be used for by exploiting modularity – a design principle that emphasizes breaking apart an overall system into simpler, well-defined parts.

Trivedi will use "recursive Markov decision processes" to describe the larger system as a collection of simpler systems, then build the overall solution to the complicated problem by combining the solutions to those simpler subtasks.

These subtasks are less energy and time-intensive to solve and the resulting modularity promotes reusability and makes explanations easier.

Through reinforcement learning that has a rigorous, formal system and that supports modularity, Trivedi's CAREER award opens a new path for complex artificial intelligence alongside neural networks, one that promises to be trustworthy, powerful and accessible to all.

Trivedi is working to democratize artificial intelligence by making machine learning more programmable, trustworthy and accessible to everyone through a prestigious CAREER award from the National Science Foundation to do so.

Traditional

White