E-commerce / Supply Chain Automation

Topics of accepted ECE NeurIPS papers include diffusion models, large language models, multi-armed bandit models, and more.

Smarter use of processor speeds saves energy without compromising training speed and performance.

The post Up to 30% of the power used to train AI is wasted. Here’s how to fix it. appeared first on Michigan Engineering News.

Some sick Black patients are likely labeled as “healthy” in AI datasets due to inequitable medical testing.

The post Accounting for bias in medical data helps prevent AI from amplifying racial disparity appeared first on Michigan Engineering News.

All have solutions, some are implemented.

The post Four election vulnerabilities uncovered by a Michigan Engineer appeared first on Michigan Engineering News.

Accepted papers for the ICML conference span topics including deep representation learning, language model fine-tuning, generative modeling, and more.

Doctoral student Can Yaras wants to reduce the carbon footprint of AI.

Doctoral student Matthew Raymond wants to facilitate the development of new and groundbreaking nanomedicines.

The DoD funds efforts to incorporate AI agents into game theory and develop microelectronics that can withstand a hot day on Venus or carry quantum information.

The post $15M for game theory with AI agents, quantum semiconductors for microelectronics and photonics  appeared first on Michigan Engineering News.

Taking advantage of the transformer neural networks that power large language models, engineers can get recipes for materials with the optical properties they need.

The post OptoGPT for improving solar cells, smart windows, telescopes and more appeared first on Michigan Engineering News.

Timekeeping in the brain is done with neurons that relax at different rates after receiving a signal; now memristors—hardware analogues of neurons—can do that too.

The post AI chips could get a sense of time appeared first on Michigan Engineering News.

The Phoenix Processor uses 30,000 times less power in sleep mode and 10 times less in active mode than comparable chips now on the market.

Award-winning research led by Prof. Qing Qu discovered an intriguing phenomenon that diffusion models consistently produce nearly identical content starting from the same noise input, regardless of model architectures or training procedures.

Current technologies already in use could help prevent crashes and deaths linked to impaired driving.

The post Auto industry deadlines loom for impaired-driver detection tech, U-M offers a low-cost solution appeared first on Michigan Engineering News.

Communities could reduce costs and cut vehicle emissions—all in the name of shortening your trip.

The post Improving traffic signal timing with a handful of connected vehicles appeared first on Michigan Engineering News.

When using only data collected before patients with sepsis received treatments or medical tests, the model’s accuracy was no better than a coin toss.

The post Widely used AI tool for early sepsis detection may be cribbing doctors’ suspicions appeared first on Michigan Engineering News.

University of Michigan researchers examine if molecular compounds in exhaled breath could lead to improved diagnosis and tracking of acute respiratory distress syndrome (ARDS).

The post U-M team receives NIH grant for collaborative research to speed ARDS diagnosis appeared first on Michigan Engineering News.

Prof. Lei Ying leads a new MURI that is focused on the interplay between online and offline networks and how they could impact disruptive behavior and events.

Regulators pinned their hopes on clinicians being able to spot flaws in explanations of an AI model’s logic, but a study suggests this isn’t a safe approach.

The post Clinicians could be fooled by biased AI, despite explanations appeared first on Michigan Engineering News.

How do chatbots based on the transformer architecture decide what to pay attention to in a conversation? They’ve made their own machine learning algorithms to tell them.

The post Understanding attention in large language models appeared first on Michigan Engineering News.

AI model that pairs text, images performs poorly on lower-income or non-Western images, potentially increasing inequality in digital technology representation.

The post Biases in large image-text AI model favor wealthier, Western perspectives appeared first on Michigan Engineering News.

Computing with a combination of light and chargeless excitons could beat heat losses and more, but excitons need new modes of transport.

The post Nextgen computing: Hard-to-move quasiparticles glide up pyramid edges appeared first on Michigan Engineering News.

And that tracks with the way our motor circuits work—we’re not that complicated.

The post Simple neural networks outperform the state-of-the-art for controlling robotic prosthetics  appeared first on Michigan Engineering News.

“I am looking forward to working with the incredible talent we have at U-M to expand the frontiers of computational science, and in more firmly establishing the role of computing in solving the grand challenge problems facing humanity.”

The post Duraisamy to lead Michigan Institute for Computational Discovery and Engineering appeared first on Michigan Engineering News.

Long aims to ensure that the information received from sensing devices is both trustworthy and confidential.

Sarabandi is recognized for his outstanding contributions to the theory and application of electromagnetics.

Roberts creates methods to better estimate the radar backscatter from corn fields, which could improve the accuracy of global biomass and soil moisture maps derived from radar observations.

The sensors will provide real-time data for smart decision-making by allowing the natural environment and the built environment to communicate seamlessly.

Using a simple and convenient touchscreen interface, the algorithm learns the assistance preferences of the wearer.

The post Choosing exoskeleton settings like a Pandora radio station appeared first on Michigan Engineering News.

Prof. Eid is looking to design the future of smart cities and infrastructures using ultra-low power wireless sensing and communications technologies.

Deep learning models that power giants like TikTok and Amazon, as well as tools like ChatGPT, could save energy without new hardware or infrastructure.

The post Optimization could cut the carbon footprint of AI training by up to 75% appeared first on Michigan Engineering News.

Led by Prof. Becky Peterson, the research focuses on a category of materials important for low power logic operations, high pixel density screens, touch screens, and haptic displays.

The post Scalable method to manufacture thin film transistors achieves ultra-clean interface for high performance, low-voltage device operation appeared first on Michigan Engineering News.

Dr. Mehdi Saligane, a leader in the open-source chip design community, was among the first researchers to fabricate a successful chip as part of Google’s multi-project wafer program.

The post Open-source hardware: a growing movement to democratize IC design appeared first on Michigan Engineering News.

Verma credits his distinguished 30-year executive career with leading technology companies, including Savi Technology, Lockheed Martin and 8×8 Inc., to a combination of education, leadership, and luck.

Known affectionately as “The Sh*tty Project,” Codling, an ECE PhD student, monitors the vibrations in pig pens to track the health of the piglets and predict when they’re in danger.

100+ researchers from across the University of Michigan and from industry gathered on North Campus for the third U-M Workshop on Data Mining.

The paper proposes an interactive natural language interface for relational databases, which enables novice users to construct complex queries.

The software enables users to ask questions about the hosts and networks that compose the Internet and get an immediate reply.

Prof. Mozafari is passionate about building large-scale data-intensive systems that are more scalable, more robust, and more predictable.

He has built software systems for information extraction, database integration, and feature engineering and applied these to problems in the social sciences.

The paper explores how automated speech recognition and crowd-sourced human correction and generation of transcripts can be traded off to improve accuracy and latency.

CSE students and faculty will collaborate as a part of a larger team to help respond to the crisis.

Secrets lurk in the dark web, the 95 percent of the internet that most of us can’t see. One U-M professor is bringing some of those secrets to light, making the digital and the real world a little safer.

The team will use fMRI to identify some of the underlying processes that occur when a code reviewer weighs in on a piece of software and its author.

The system can add more flexibility to task management apps to help learning users make informed decisions about their time.

K-5 teachers and students throughout Michigan are building thriving learning communities online by using free deeply-digital, standards-aligned curricula and platform developed by the U-M Center for Digital Curricula.

Five multidisciplinary research teams are working on projects to assist with the coronavirus outbreak and to help find solutions to pressing problems.

All of the research being presented focuses on getting the absolute best performance from the tiniest circuits, sensors, and electronic devices.

Millimeter-sized computers log the temperature and pressure from deep within oil wells.

The Michigan Daily profiles Professors David Blaauw and Dennis Sylvester, who are this year’s recipients of the 2019 Distinguished University Innovator Award.

Blaauw’s innovations in low-power computing led to development of the Michigan Micro Mote, the world’s smallest computer.

Xiang’s research focuses on developing new methods to synthesize different control and sensing strategies in a discrete-event system.

Hero is honored for his extraordinary accomplishments that have brought distinction to himself, his students, and to the entire University.

This platform has enabled a variety of sensors that can fit inside the human body, made possible by several breakthroughs in ultra-low power computing.

Mohammad has developed a new way to remotely measure the thickness of ice and snow with a technology he calls wideband autocorrelation radiometry (WiBAR).

The AAAS seeks to advance science, engineering, and innovation throughout the world for the benefit of all people.

ECE alum Rick Bergman, CEO of Synaptics, is working to make tomorrow’s technology user friendly, safe, and reliable. The company hopes to lead what they call “the human interface revolution.”

The paper outlines a better way to quantify forest structure, which has been successful in two tree species.

Keenan Rebara hopes to add to the fun of spinning the Cube using his a bit of physics and sensors.

A solar cell combined with a camera sensor collects photons to provide electricity.

An award-winning method will help us better understand how much snow is on the ground.

Applications include managing large networked systems, such as sensor networks, power grids, or computer networks.

Huang won the Best Student Paper Award at the IEEE International Conference on Computational Electromagnetics for her work developing better electromagnetic models that calculate microwave interactions with tree and vegetation cover.

Electrify hosted its first Detroit Tech Camp at the Michigan Engineering Zone this summer to give Detroit-area students greater access to engage with science, technology, engineering, and mathematics.

Wu is working on advanced metasurfaces, which could help next-generation wireless communication, commercial and military radar systems, imaging, and antenna systems.

PhD student Ester Bentley designs smaller, better 3D mechanical resonators for use in high-performance gyroscopes to help unmanned systems navigate when GPS signal is jammed or lost.

The WAND wireless sensor developed in a collaboration between Total, an oil & gas company, and the University of Michigan is revolutionizing well monitoring

Students say Ulaby, a member of the National Academy of Engineering and recipient of the Edison Medal, is one of the best professors – and people – they’ve ever known.

Cheng-Hsun Lu,
Shih-Chi Liao, and Jiale Zhang have been awarded the Rackham International Students Fellowship/Chia-Lun Lo Fellowship.

The system could enable new biomedical implants as well as home-, building- and bridge-monitoring devices.

Until now, ubiquitous computing has been hampered by the size of necessary batteries—but Ambiq Micro is changing that, with their energy-efficient micro-controllers.

U-M faculty have developed what is believed to be the first complete millimeter-scale computing system, with applications in radio communication and wireless sensing.

This research is expected to have a fundamental and long term impact on a diverse set of applications ranging from energy conservation to health care.

Wentzloff aims to remove the necessity of a power outlet or even a battery to power miniature sensors.

Avish is currently conducting research on ultra-low power radio technology and designing a low-power RF power amplifier.

Movellus Circuits’ product is a patent-pending clock generator technology that is smaller, cheaper, and faster than existing solutions.

The chips’ extreme energy efficiency enables them to be powered without a battery from harvested energy sources like vibration, thermal gradients, and more.

Instead of a battery, the chip Nathan is engineering uses two solar cells that look like they belong on a calculator.

Dr. Hanson is the co-founder of a startup semiconductor company that plans to lead the low-power revolution in electronics by powering the Internet of Things.

At the age of 24, Yang sold his company ChinaRen for $35 million.

A brief history of what led to the technical feat known as the Michigan Micro Mote, a tiny speck of a computer that does it all.

Shahin and Sassan discussed everything from the acquisition trends of small vs. large companies to the importance of building a team with a range of expertise.

Ambiq Micro, Crossbar, Inc., and PsiKick, are leading the way in ultra-low power chip design, pioneering computer memory, and ultra-low power wireless sensor platforms.

IoT applications are the next wave of computing and the next driving force of the semiconductor industry. The startup PsiKick [now Everactive] is helping shape this future.

Cyber-physical systems are smart, networked systems with embedded sensors, processors, and actuators that are designed to interact with the physical world.

The M3 is a fully autonomous computing system that acts as a smart sensing system.

Avish conducts research on ultra-low power and battery-less integrated circuits.

Adkins plans to continue his graduate studies in electrical engineering and computer science at the University of California Berkeley.

Movellus Circuits won $25,000 in the University Research Highlight and People’s Choice categories

Professors Blaauw and Sylvester showcase capabilities of tiny computing

Kim’s research is expected to impact the future design and wireless operation of the next generation of Internet of Things (IoT) devices

The law of small numbers could impact the next generation of tools that deal with data.

Keeping time in the Internet of Things with frequency scaling

Researchers built the first millimeter-scale transmitter and antenna that can talk Bluetooth Low Energy with ease.

The Michigan Micro Mote gets a new gallium arsenide solar cell for added power and adaptability.

With the help of two NSF awards totaling $1.7m, Prof. Hessam Mahdavifar is tackling new problems to improve the reliability of communication systems for 5G and beyond.

Prof. David Wentzloff’s paper examining the trends and techniques to achieve ultra-low power receivers was honored by the IEEE Custom Integrated Circuits Conference

Ryan is an electrical engineering undergrad interested in military systems and devices.

PhD candidate Mohammad Vahid Jamali won a Best Paper award at IEEE ICC for his work on Product AutoEncoders, which could help shape future generations of wireless networks, IoT, and autonomous systems.

With a radio specifically designed to communicate through tissue, researchers from the Electrical and Computer Engineering are adding another level to a computer platform small enough to fit inside a medical grade syringe.

Some believed early Michigan brain researchers were engaging in “science fiction” – until development of an advanced tool for forging breakthroughs proved them wrong.

The post The Michigan Probe: Changing the Course of Brain Research appeared first on Michigan Engineering News.

Prof. Todd Coleman’s group is tackling the challenging problem of getting high-fidelity monitoring to work affordably at home.

SPARC awarded $1M to a U-M project developing better nerve mapping.

UM-SEDS co-President Arun Nagpal develops ENG 100 section to expose freshman to space science and atmospheric sensing.

Prof. Tsang is a world-renowned expert in the field of theoretical and computational electromagnetics, and in particular microwave remote sensing of the earth.

The satellite mission to collect global data of surface soil moisture can help weather forecasting around the world.

An award-winning modeling method will help us better understand our natural environment

To dial in on exact wind speeds, researchers needed to reverse engineering the signals from satellites.

For the 50th anniversary of the Apollo 11 moon landing, U-M ECE takes a look back – and a look forward – to how our professors, students, and alums have made their mark on the field.

England has dedicated more than two decades of his distinguished career helping students reach for the stars to understand more about Earth and other planets.

Yektakhah’s system improves on the speed, portability, and accuracy of many commercial models

With $7.5M MURI grant, Professor Anthony Grbic is developing metamaterials for a new generation of integrated electromagnetic and photonic systems.

Liu’s most recent research involves online learning, modeling of large-scale internet measurement data, and incentive mechanisms for security games.

DARPA is trying to build a system that can turn large data sets into models that can make predictions, and U-M is in on the project.

Mingyan Liu, recipient of the 2018 Distinguished Innovator of the Year award, gave a talk about her startup company and participated on a panel discussing data science commercialiation.

Hero and Lindquist took a few minutes to talk about the impact of machine learning on Signal Processing and Control Systems, and what they plan to do about it

Kim takes an interdisciplinary approach to tackle challenges in heterogeneous classes of energy-efficient and versatile communication systems.

His research develops computational methods for learning succinct representations from high-dimensional data.

Prof. Al Hero was interviewed and gave a presentation about his research using machine learning to improve our understanding of the human gut

Komma, a PhD student, is working to develop robust low powered localization technology for Artificial Intelligence enabled Internet of Things in locations where GPS is limited or blocked.

In a project he calls the “Marauder’s Map,” Prof. Zhang uses machine learning-based data models, physics models, and heuristic models to turn physical structures into sensing devices.

The study yields new insights into the survival of a native snail important to Tahitian culture and ecology and to biologists studying evolution, while proving the viability of similar studies of very small animals including insects

A longstanding collaboration between engineers and neuroscientists leads to new insights into how neurons work in the hippocampus.

Centralizing available data in the intelligent systems community through a COmputer Vision Exchange for Data, Annotations and Tools, called COVE.

Predicting future disasters is an important goal of those participating in the IEEE Geoscience and Remote Sensing Symposium

A professor of electrical engineering and computer science is awarded one of engineering’s top honors.

The College of Engineering honors ECE PhD candidate Tianlin Wang for his excellent research in remote sensing as well as his leadership and service to the community.

Prof. Yogesh Gianchandani and Dr. Yutao Qin received an “Outstanding Paper Award” for their fully electronic micro gas chromatography system.

In a project funded by National Geographic, ECE researchers are teaming up with the department of Ecology and Evolutionary Biology to advance our understanding of monarch butterfly migration with the most ambitious iteration of the Michigan Micro Mote yet.

Roberts works to improve remote sensing of soil moisture, which is important for environmental conservation, natural resource management, and agriculture.

Electrical Engineering undergrad Madeline Evans is a key researcher on a project that uses NASA’s Cyclone Global Navigation Satellite System to monitor microplastic pollution that harms marine ecosystems.

Using retired electric vehicle batteries, the project plans to enable widespread and equitable access to sustainable power and energy through sustainable energy storage.

‘You shouldn’t need a Ph.D. to design new computing systems.’

The latest from IBM and now the University of Michigan is redefining what counts as a computer at the microscale.

How changing the rules of computing could lighten Big Data’s impact on the internet.

PhD student Trevor Odelberg is looking to enable long range, highly reliable, and low-power cellular IoT devices that one day can run entirely on harvested energy, reducing battery waste and empowering devices to last for decades.

The post Batteryless next-generation cellular devices could empower a more sustainable future appeared first on Michigan Engineering News.

The frame in which a human marks out the boundaries of an object makes a huge difference in how well AI software can identify that object through the rest of the video.

‘Mcity 2.0’ will give researchers, many without testing resources, remote access to the Mcity Test Facility—creating a more equitable playing field in mobility.

The post $5M to enable remote, next-generation autonomous vehicle testing at Mcity appeared first on Michigan Engineering News.

Surprise findings could upend the current drug discovery approach for treating one of the most dangerous hospital-borne infections.

The post A ‘decathlon’ for antibiotics puts them through more realistic testing appeared first on Michigan Engineering News.

Study sheds light on the future of the massive Thwaites Glacier.

The post ‘Doomsday Glacier’ may be more stable than initially feared appeared first on Michigan Engineering News.

Predictive modeling could help power companies get more consistent output from renewables.

The post Making wind power more predictable: A Q&A with Eunshin Byon appeared first on Michigan Engineering News.

Engineering assistant professor Raed Al Kontar outlines a new paradigm for connected devices.

The post Keeping the world connected, without sacrificing privacy appeared first on Michigan Engineering News.

“Acoustic fields are unexpectedly richer in information than is typically thought.”

The post Mining soundwaves: Researchers unlock new data in sonar signals appeared first on Michigan Engineering News.

The algorithm can pick out weak signals from nuclear weapons materials, hidden in ordinary radiation sources like fertilizer.

U-M Aerospace Engineering Professor Venkat Raman advocates for more versatile and powerful modeling tools to meet computational demands of next-generation aircraft design.

The post Opinion: Future aerospace enterprises will demand more advanced modeling and simulation appeared first on Michigan Engineering News.

The finding could have implications for future agile autonomous aerial vehicles.

The post Avian secret: The key to agile bird flight is switching quickly between stable and unstable gliding appeared first on Michigan Engineering News.

With new courses at the UG and graduate level, ECE is delivering state-of-the-art instruction in machine learning for students in ECE, and across the University

Public policy and engineering team up to improve food access.

The post Hunger and COVID: Fighting pandemic-related food insecurity in Detroit appeared first on Michigan Engineering News.

Microsoft-supported project to coordinate site locations, supply distribution.

The post How big data could optimize COVID-19 testing appeared first on Michigan Engineering News.

A conventional approach to HIV vaccination does not induce immune responses in everyone equally, and a new computer model shows why.

The post Predicting how well a vaccine will work for you appeared first on Michigan Engineering News.

The model can be used as a tool to inform decision-makers and individuals on relative risks and advantages associated with a layered defense.

The post Michigan Engineering group creates model for layering COVID-19 defenses appeared first on Michigan Engineering News.

An electrode array implanted in the brain predicts finger motions in near real time.

The post Individual finger control for advanced prostheses demonstrated in primates appeared first on Michigan Engineering News.

Built to handle falls, and with two extra motors in each leg, the new robot will help U-M roboticists take independent robotic walking to a whole new level.

The post Latest two-legged walking robot arrives at Michigan appeared first on Michigan Engineering News.

Data gleaned from cameras and sensors increases predictive accuracy.

The post Teaching self-driving cars to predict pedestrian movement appeared first on Michigan Engineering News.

U-M startup says robotic food deliveries have quadrupled.

The post Delivery robots help Ann Arbor restaurants weather COVID appeared first on Michigan Engineering News.

An open-source perception and movement system, to be developed with NSF funding, could enable robots that partner with humans in fires and disaster areas.

The post $1M for open-source first-responder robots appeared first on Michigan Engineering News.

Tested without needing hospitals to share data, the method for developing the model could speed further improvements in medical prediction tools.

The post Open-source patient model tops industry standard appeared first on Michigan Engineering News.

The adaptive immune system serves as a template for defending neural nets from confusion-sowing attacks

Quantum materials emit light as though it were only a positive pulse, rather than a positive-negative oscillation.

The post Emulating impossible “unipolar” laser pulses paves the way for processing quantum information appeared first on Michigan Engineering News.

The new computer model accurately predicts the behavior of millions of microbial communities from hundreds of experiments, an advance toward precision medicine.

Virtual assortment of user devices provides a realistic training environment for distributed machine learning, protects privacy by learning where data lives.

The post Open source platform enables research on privacy-preserving machine learning appeared first on Michigan Engineering News.

New technique could enable processing speeds a million to a billion times faster than today’s computers and spur progress in many-body physics.

The post Seeing electron movement at fastest speed ever could help unlock next-level quantum computing appeared first on Michigan Engineering News.