News

CU �鶹ӰԺ receives $1 million grant to fund minority PhD STEM students

Anonymous — Thu, 25 Aug 2022 20:54:10 +0000

CU �鶹ӰԺ receives $1 million grant to fund minority PhD STEM students Anonymous (not verified) Thu, 08/25/2022 - 14:54

Grace Wilson

The fellowship will provide tuition support, a $34,000 stipend for two years and a unique support system for 12 students to advance their PhD studies and research in computer science and other fields.

Off

Traditional

White

PhD students explore trust between humans and robots with augmented reality version of Minesweeper

Anonymous — Mon, 25 Jul 2022 21:55:09 +0000

PhD students explore trust between humans and robots with augmented reality version of Minesweeper Anonymous (not verified) Mon, 07/25/2022 - 15:55

Ellen Fike

When you should trust your robot teammate? Computer science PhD students Aaquib Tabrez and Matthew Luebbers, along with their advisor Assistant Professor Bradley Hayes, used augmented reality minesweeper to gain insight into a robot’s decision-making process and were awarded runner-up for best student paper at AAMAS 2022.

Off

Traditional

White

How to turn throwaway cardboard into a DIY arcade game

Anonymous — Fri, 22 Jul 2022 15:48:09 +0000

How to turn throwaway cardboard into a DIY arcade game Anonymous (not verified) Fri, 07/22/2022 - 09:48

Daniel Strain

Tinycade, the brainchild of Gyory and his colleagues at ATLAS, brings a do-it-yourself spirit to the world of video games. Tinycade allows anyone, anywhere to make a working arcade machine that can fit on a side table, with just a smartphone, some cardboard, two small mirrors and bric-a-brac like rubber bands and toothpicks.

Off

Traditional

White

CU �鶹ӰԺ researchers explore the engineering of bee honeycombs

Anonymous — Tue, 19 Jul 2022 19:20:43 +0000

CU �鶹ӰԺ researchers explore the engineering of bee honeycombs Anonymous (not verified) Tue, 07/19/2022 - 13:20

Jeff Zhender

Assistant professor Orit Peleg in the Department of Computer Science will work on a new $497,000 grant with aerospace assistant professor López Jiménez Ann to explore how bees build honeycombs, research that supports bio-inspired system designs in swarm robotics and lightweight cellular structures.

Off

Traditional

White

$20M ESIIL center aims to foster a “revolution” in environmental data science

Anonymous — Mon, 11 Jul 2022 18:45:58 +0000

$20M ESIIL center aims to foster a “revolution” in environmental data science Anonymous (not verified) Mon, 07/11/2022 - 12:45

Computer Science Associate Professor Claire Monteleoni is the AI/Machine Learning Lead for a new $20M NSF Center at CU �鶹ӰԺ, housed in CIRES, the Cooperative Institute for Research in Environmental Sciences. The new center is a major new data science and diversity effort including partners from multiple institutions around the world.

Off

Traditional

White

CU �鶹ӰԺ part of $5.8M grant to improve cyber-physical transportation systems

Anonymous — Wed, 29 Jun 2022 06:00:00 +0000

CU �鶹ӰԺ part of $5.8M grant to improve cyber-physical transportation systems Anonymous (not verified) Wed, 06/29/2022 - 00:00

CS assistant professor Majid Zamani and his team are part of a new $5.8M grant from the NSF to help build the intelligent transportation systems of the future.

Off

Traditional

White

Helping robots recover from failure: a Q+A with NASA graduate fellowship winner Gilberto Briscoe-Martinez

Anonymous — Mon, 27 Jun 2022 22:39:15 +0000

Helping robots recover from failure: a Q+A with NASA graduate fellowship winner Gilberto Briscoe-Martinez Anonymous (not verified) Mon, 06/27/2022 - 16:39

Grace Wilson

We all fail sometimes, but how we respond to those failures is important. Classical robotic failure mitigation focuses on trying to anticipate every way a system could fail and having an answer.

Gilberto Briscoe-martinez (PhDCompSci‘26) wants robots to be able to learn from their mistakes and continue persevering to finish a task as best they can. He was recently awarded a NASA Space Technology Graduate Research (NSTGRO) fellowship for his exploration of this topic, titled “Enabling Long-term Robot Autonomy through Adaptable Fault Resilience”. He is the first Computer Science P.h.D. student at CU �鶹ӰԺ to receive this award. The fellowship will fund Gilberto for 4 years and will allow him to collaborate with scientists at NASA research centers.

Alessandro Roncone, Briscoe-martinez' advisor and director of the Human Interaction and RObotics (HIRO) Group at CU �鶹ӰԺ, said "The NASA NSTGRO is one of the most prestigious fellowships awarded to graduate students in robotics at the national level. With this support from NASA, Gilberto will be able to not only advance the science of robotic fault resilience, but also understand how to transfer his discoveries to real systems, and, eventually, see his research deployed in space."

We asked Briscoe-martinez a few questions about his research and fellowship, here are his answers:

How would you describe this project to someone not familiar with your area of research?

Humans have an incredible, innate ability to compensate for an injury that we experience. If you’ve ever stubbed your toe, you’ve probably tried not to put weight on it without explicitly thinking about doing so. On the other hand, robots have an unchanging understanding of their physical selves. If anything breaks, even something as small as a piece of the robot’s gripper falling off, the robot is useless until a human can fix it. For NASA, this presents a problem. Robots will need to work reliably for months or even years on new space stations and future inter-planetary missions with few or no ways to fix themselves. My research will discover methods for these robots to compensate for injuries and wear they experience, as humans do, so they can continue to do their mission-critical work.

What does winning this fellowship mean to you?

For me, receiving this fellowship feels like a transition point. Until now, I’ve only used robotic technology developed by others, solving “engineering problems,” as my advisor likes to say. Now, I look forward to researching the ways that will push robotic abilities to the stars. The research I will be conducting through this fellowship is critical to enabling humans to go beyond the reaches of our planet. I am beyond excited to be at the forefront of space robotics.

How has it been working in the Human Interaction and RObotics (HIRO) Group?

My experience in the HIRO lab has been fantastic. The lab, as a whole, is focused on wholistically researching robotic systems. This enables unique collaboration opportunities between those working on the main research threads of robotic ability, physical Human-Robot Interaction(HRI), and Social HRI. My colleagues have been great inspirations because we have different outlooks, experiences, and research approaches. In addition, I believe the lab has a great work-life balance, where we are pushed to excel but can go out to ski on a snow day in the middle of the week.

What inspired your love of robotics?

My curiosity was first piqued when I took a middle-school robotics class after my school received a grant from LEGO to use their Mindstorm systems. I realized robotics would be a lifelong passion when I joined my high school robotics club. We competed in both the VEX and FIRST robotics competitions. By designing and building robots to complete many challenges, I realized that the only limits of robotic ability are the limits of our imagination. From those experiences, I was inspired to pursue the robotics research that I do today.

Any advice for folks not sure if STEM is right for them?

My advice for someone on the fence about entering STEM is to not be afraid to try it. If you’re thinking about it then that curiosity is already there which can turn into your life’s work. And it is important to remember that if it turns out that STEM is not right for you, switching majors, and even switching careers is something that many people successfully do.

Briscoe-martinez was recently awarded a NASA Space Technology Graduate Research (NSTGRO) fellowship. He is the first Computer Science P.h.D. student at CU �鶹ӰԺ to receive this award, which will provide him with 4 years of research funding and NASA collaboration in enabling long-term robot autonomy through adaptable fault resilience.

Off

Traditional

White

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.

Off

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.

Off

Traditional

White

Keeping water on the radar: Machine learning to aid in essential water cycle measurement

Anonymous — Fri, 20 May 2022 17:45:44 +0000

Keeping water on the radar: Machine learning to aid in essential water cycle measurement Anonymous (not verified) Fri, 05/20/2022 - 11:45

Grace Wilson

Department of Computer Science assistant professor Chris Heckman and CIRES research hydrologist Toby Minear have been awarded a Grand Challenge Research & Innovation Seed Grant to create an instrument that could revolutionize our understanding of the amount of water in our rivers, lakes, wetlands and coastal areas by greatly increasing the places where we measure it.

The new low-cost instrument would use radar and machine learning to quickly and safely measure water levels in a variety of scenarios.

This work could prove vital as the USDA recently proclaimed the entire state of Colorado to be a "primary natural disaster area" due to an ongoing drought that has made the American West potentially the driest it has been in over a millennium. Other climate records across the globe also continue to be broken, year after year. Our understanding of the changing water cycle has never been more essential at a local, national and global level.

A fundamental part to developing this understanding is knowing changes in the surface height of bodies of water. Currently, measuring changing water surface levels involves high-cost sensors that are easily damaged by floods, difficult to install and time consuming to maintain.

"One of the big issues is that we have limited locations where we take measurements of surface water heights," Minear said.

A new method

Heckman and Minear are aiming to change this by building a low-cost instrument that doesn't need to be in a body of water to read its average water surface level. It can instead be placed several meters away – safely elevated from floods.

The instrument, roughly the size of two credit-cards stacked on one another, relies on high-frequency radio waves, often referred to as "millimeter wave", which have only been made commercially accessible in the last decade.

Through radar, these short waves can be used to measure the distance between the sensor and the surface of a body of water with great specificity. As the water's surface level increases or decreases over time, the distance between the sensor and the water's surface level changes.

The instrument's small form-factor and potential off-the-shelf usability separate it from previous efforts to identify water through radar.

It also streamlines data transmitted over often limited and expensive cellular and satellite networks, lowering the cost.

In addition, the instrument will use machine learning to determine whether a change in measurements could be a temporary outlier, like a bird swimming by, and whether or not a surface is liquid water.

Machine learning is a form of data analysis that seeks to identify patterns from data to make decisions with little human intervention.

While traditionally radar has been used to detect solid objects, liquids require different considerations to avoid being misidentified. Heckman believes that traditional ways of processing radar may not be enough to measure liquid surfaces at such close proximity.

"We're considering moving further up the radar processing chain and reconsidering how some of these algorithms have been developed in light of new techniques in this kind of signal processing," Heckman said.

Citizen science

In addition to possible fundamental shifts in radar processing, the project could empower communities of citizen scientists, according to Minear.

"Right now, many of the systems that we use need an expert installer. Our idea is to internalize some of those expert decisions, which takes out a lot of the cost and makes this instrument more friendly to a citizen science approach," he said.

By lowering the barrier of entry to water surface level measurement through low-cost devices with smaller data requirements, the researchers broaden opportunities for communities, even in areas with limited cellular networks, to measure their own water sources.

The team is also committing to open-source principles to ensure that anyone can use and build on the technology, allowing for new innovations to happen more quickly and democratically.

Broader applications

Minear, who is a Science Team and Cal/Val Team member for the upcoming NASA Surface Water and Ocean Topography (SWOT) Mission, also hopes that the new instrument could help check the accuracy of water surface level measurements made by satellites.

These sensors could also give local, regional and national communities more insight into their water usage and supply over time and could be used to help make evidence-informed policy decisions about water rights and usage.

"I'm very excited about the opportunities that are presented by getting data in places that we don't currently get it. I anticipate that this could give us better insight into what is happening with our water sources, even in our backyard," said Heckman.

Off

Traditional

White