research

image of Chelsea Finn
July 2021

Assistant Professor Chelsea Finn and her team designed a neural network system solely for Stanford's programming class. They used techniques that could automate student feedback in other situations, including for classes beyond programming.

Chelsea's system spent hours analyzing examples from old midterms, learning from a decade of possibilities. Then it was ready to learn more. When given just a handful of extra examples from the new exam offered this spring, it could quickly grasp the task at hand.

"It sees many kinds of problems," said PhD candidate Mike Wu. "Then it can adapt to problems it has never seen before."

This spring, the system provided 16,000 pieces of feedback, and students agreed with the feedback 97.9 percent of the time, according to a study by the researchers. By comparison, students agreed with the feedback from human instructors 96.7 percent of the time.

 

Chelsea Finn is an assistant professor of electrical engineering and computer science.

 

 

Excerpted from The New York Times, "Can A.I. Grade Your Next Test?"

Related News

image of prof Dorsa Sadigh
July 2021

Congratulations to Assistant Professor Dorsa Sadigh! She is included with MIT's 35 Innovators Under 35.

By developing new ways for computers to anticipate people's actions, Dorsa Sadigh wants to help pave the way for a future in which human and robots do things like share the roads. Her research interests lie at the intersection of robotics, machine learning, and control theory. Dorsa's research group develops algorithms for AI agents that safely and reliably interact with people. Her research group is ILIAD

The MIT Technology Review, 35 Innovators Under 35 looks at where technology is now, and where it's going and who's taking it there. Congratulations to all the innovators!

 

Please join us in congratulating Dorsa on this recognition!

 

 

Related News

 

image of prof Howard Zebker
June 2021

Venus is often called Earth's sister planet or twin because the two worlds are of similar size and density. Yet the second rock from the sun is hot and inhospitable in the extreme. "That's one of the reasons we know so little about the surface," said Professor Howard Zebker. "If you send a spacecraft to the surface of Venus, which has been done several times, they only last a few minutes until the hot acid burns them up."

Howard is a member of the science team for VERITAS, one of three missions to Venus announced in June 2021 by NASA and the European Space Agency. As part of the VERITAS mission – which is expected to launch around 2028-2030 – instruments aboard the spacecraft will measure how long it takes radar signals to bounce back from a series of precise locations at different times. This will yield pairs of images that can be combined to reveal changes in altitude at the surface using a technique known as interferometric synthetic aperture radar, or InSAR.

Algorithms and techniques pioneered by Howard will help to guide these measurements and translate them into high-resolution 3D maps of any ongoing deformation of Venus' outermost layer. On Earth, InSAR has been used to map uplift and subsidence related to groundwater pumping; to detect sinkholes; and to study glacier movements, earthquakes, volcanic eruptions, landslides and more. But this is the first time the techniques will be used by spacecraft to identify active fault movements beyond our world.

While NASA's Jet Propulsion Laboratory in Pasadena, Calif. will manage the VERITAS (Venus Emissivity, Radio Science, InSAR, Topography & Spectroscopy) mission, students working in Professor Zebker's lab – Radar Remote Sensing & Radar Interferometry Group will help to refine algorithms for the mission over the next several years and work to interpret the data that come in once VERITAS makes it into orbit.

Howard Zebker is a professor of geophysics and electrical engineering. He discusses his role in the VERITAS mission; how InSAR will help to answer key questions about volcanic activity and tectonic plates on Venus; why our hothouse twin may hold insights relevant to modeling of climate change on our own planet; and paths for interested students to get involved.

 

Excerpted from Stanford News, "Is Venus still geologically active? Stanford expert explains technology powering NASA's quest to understand Earth's twin", June 29, 2021.

Related News

image of professor Eric Pop
June 2021

Professor Eric Pop and team describe the ability to produce nanoscale flexible electronics In their paper, "High-Performance Flexible Nanoscale Transistors Based on Transition Metal Dichalcogenides," published in Nature Electronics. Flexible electronics promise bendable, shapeable, yet energy-efficient computer circuits that can be worn on or implanted in the human body to perform myriad health-related tasks. Future variations future of the circuits will communicate wirelessly with the outside world – another large leap toward viability for flextronics, particularly those implanted in the human body or integrated deep within other devices connected to the internet of things.

[...]
With a prototype and patent application complete, postdoc Alwin Daus and Professor Eric Pop have moved on to their next challenges of refining the devices. They have built similar transistors using two other atomically thin semiconductors (MoSe2 and WSe2) to demonstrate the broad applicability of the technique.

Meanwhile, Alwin said that he is looking into integrating radio circuitry with the devices, which will allow future variations to communicate wirelessly with the outside world – another large leap toward viability for flextronics, particularly those implanted in the human body or integrated deep within other devices connected to the internet of things.

Eric reports, "This is more than a promising production technique. We've achieved flexibility, density, high performance and low power – all at the same time. This work will hopefully move the technology forward on several levels."

Co-authors include postdoctoral scholars Sam Vaziri and Kevin Brenner, EE doctoral candidates Victoria Chen, Çağıl Köroğlu, Ryan Grady, Connor Bailey and Kirstin Schauble, and research scientist Hye Ryoung Lee. Pop Lab People

 

Excerpted from "Stanford researchers develop new manufacturing technique for flexible electronics" Stanford News

image of Londa Schiebinger and James Zou
June 2021

Debiasing artificial intelligence (AI)

In the medical field, AI encompasses a suite of technologies that can help diagnose patients’ ailments, improve health care delivery and enhance basic research. The technologies involve algorithms, or instructions, run by software. These algorithms can act like an extra set of eyes perusing lab tests and radiological images; for instance, by parsing CT scans for particular shapes and color densities that could indicate disease or injury.

Problems of bias can emerge, however, at various stages of these devices’ development and deployment, James explained. One major factor is that the data for forming models used by algorithms as baselines can come from nonrepresentative patient datasets.

By failing to properly take race, sex and socioeconomic status into account, these models can be poor predictors for certain groups. To make matters worse, clinicians might lack any awareness of AI medical devices potentially producing skewed results. 


In a new perspective paper, James Zou and Londa Schiebinger discuss sex, gender and race bias in medicine and how these biases could be perpetuated by AI devices. 
 
James and Londa suggest several short- and long-term approaches to prevent AI-related bias, such as changing policies at medical funding agencies and scientific publications to ensure the data collected for studies are diverse, and incorporating more social, cultural and ethical awareness into university curricula.

“The white body and the male body have long been the norm in medicine guiding drug discovery, treatment and standards of care, so it’s important that we do not let AI devices fall into that historical pattern,” said Londa Schiebinger, the John L. Hinds Professor in the History of Science in the School of Humanities and Sciences and senior author of the paper published in the journal EBioMedicine.

“As we’re developing AI technologies for health care, we want to make sure these technologies have broad benefits for diverse demographics and populations,” said James Zou, assistant professor of biomedical data science and, by courtesy, of computer science and of electrical engineering and co-author of the study.

The matter of bias will only become more important as personalized, precision medicine grows in the coming years, said the researchers. Personalized medicine, which is tailored to each patient based on factors such as their demographics and genetics, is vulnerable to inequity if AI medical devices cannot adequately account for individuals’ differences.

“We’re hoping to engage the AI biomedical community in preventing bias and creating equity in the initial design of research, rather than having to fix things after the fact,” said Londa Schiebinger.
 
 
 
 

Related News

image of prof Boris Murmann
May 2021

Congratulations to Professor Boris Murmann for receiving the 2021 Semiconductor Industry Association (SIA)-SRC University Researcher Award. Award winners are selected by SRC members who truly recognize great contributions to the semiconductor industry.

Source: src.org/award/university-researcher/

The SIA University Research Award was established in 1995 by the Semiconductor Industry Association (SIA) to recognize lifetime research contributions to the U.S. semiconductor industry by university faculty.

Please join us in recognizing Boris for his significant contributions in the semiconductor industry.

 

Related News

 

 

image of prof Subhasish Mitra
May 2021

Congratulations to Professor Subhasish Mitra for receiving the 2021 Semiconductor Industry Association (SIA)-SRC University Researcher Award. Award winners are selected by SRC members who truly recognize great contributions to the semiconductor industry.

Please join us in recognizing Subhasish for his significant contributions in the semiconductor industry.

 

The SIA University Research Award was established in 1995 by the Semiconductor Industry Association (SIA) to recognize lifetime research contributions to the U.S. semiconductor industry by university faculty.

Source: SRC.org/award/university-researcher

 

Related

 

image of prof Eric Pop
May 2021

A team of Stanford researchers including EE Professor Eric Pop report the design and fabrication of single-wall carbon nanotube thermoelectric devices on flexible polyimide substrates as a basis for wearable energy converters.

source: ScienceDaily.com [...]
Inspiration came from a desire to ultimately fabricate energy converting devices from the same materials as the active devices themselves, so they can blend in as an integral part of the total system. Today, many biomedical nanodevices' power supplies come from several types of batteries that must be separated from the active portion of the systems, which is not ideal.

In Applied Physics Letters, the researchers report the design and fabrication of single-wall carbon nanotube thermoelectric devices on flexible polyimide substrates as a basis for wearable energy converters.

"Carbon nanotubes are one-dimensional materials, known for good thermoelectric properties, which mean developing a voltage across them in a temperature gradient," said Professor Eric Pop. "The challenge is that carbon nanotubes also have high thermal conductivity, meaning it's difficult to maintain a thermal gradient across them, and they have been hard to assemble them into thermoelectric generators at low cost."

The group uses printed carbon nanotube networks to tackle both challenges.

Professor Pop continued, "For example, carbon nanotube spaghetti networks have much lower thermal conductivity than carbon nanotubes taken alone, due to the presence of junctions in the networks, which block heat flow. Also, direct printing such carbon nanotube networks can significantly reduce their cost when they are scaled up."

Thermoelectric devices generate electric power locally "by reusing waste heat from personal devices, appliances, vehicles, commercial and industrial processes, computer servers, time-varying solar illumination, and even the human body," said Hye Ryoung Lee, lead author and a research scientist.

"To eliminate hindrances to large-scale application of thermoelectric materials – toxicity, materials scarcity, mechanical brittleness – carbon nanotubes offer an excellent alternative to other commonly used materials," Lee said.

The group's approach demonstrates a path to using carbon nanotubes with printable electrodes on flexible polymer substrates in a process anticipated to be economical for large-volume manufacturing. It is also "greener" than other processes, because water is used as the solvent and additional dopants are avoided.

Excerpted from "Nontoxic, flexible energy converters could power wearable devices" April 27, 2021

 

Related

image of PhD candidate Riley Culberg
April 2021

Research by EE PhD candidate Riley Culberg and Prof. Dustin Schroeder is revealing the long-term impact of vast ice melt in the Arctic.

Using a new approach to ice-penetrating radar data, researchers show that this melting left behind a contiguous layer of refrozen ice inside the snowpack, including near the middle of the ice sheet where surface melting is usually minimal. Most importantly, the formation of the melt layer changed the ice sheet's behavior by reducing its ability to store future meltwater. The research appears in Nature Communications.

"When you have these extreme, one-off melt years, it's not just adding more to Greenland's contribution to sea-level rise in that year – it's also creating these persistent structural changes in the ice sheet itself," said lead author Riley Culberg, EE PhD candidate. "This continental-scale picture helps us understand what kind of melt and snow conditions allowed this layer to form."

Airborne radar data, a major expansion to single-site field observations on the icy poles, is typically used to study the bottom of the ice sheet. But by pushing past technical and computational limitations through advanced modeling, the team was able to reanalyze radar data collected by flights from NASA's Operation IceBridge from 2012 to 2017 to interpret melt near the surface of the ice sheet, at a depth up to about 50 feet.

"Once those challenges were overcome, all of a sudden, we started seeing meltwater ice layers near the surface of the ice sheet," EE courtesy professor, Dustin Schroeder said. "It turns out we've been building records that, as a community, we didn't fully realize we were making."

Melting ice sheets and glaciers are the biggest contributors to sea-level rise – and the most complex elements to incorporate into climate model projections. Ice sheet regions that haven't experienced extreme melt can store meltwater in the upper 150 feet, thereby preventing it from flowing into the ocean. A melt layer like the one from 2012 can reduce the storage capacity to about 15 feet in some parts of the Greenland Ice Sheet, according to the research.

 

 

Excerpted from "Stanford researchers reveal the long-term impacts of extreme melt on Greenland Ice Sheet", Stanford News, April 20, 2021

image of prof Shanhui Fan
April 2021

Professor Shanhui Fan presented his latest advances in radiative cooling at annual energy sector conference. Shanhui's radiative cooling harvests electricity from the coldness of the universe, which in turn, can be harvested on Earth for several renewable energy applications. For millennia, humans in regions where the ambient temperature never falls below freezing have used the concept to make ice by burying water at night.

Radiative cooling could have a significant impact on lowering electricity use and boosting output of renewables, but it will require advances in blackbody emitters, materials that absorb heat and radiate the heat at frequencies that send it into space.

"This requires a good blackbody emitter," said Shanhui, "but we can cool objects to a temperature 13 degrees Celsius (55 degrees Fahrenheit) below the ambient temperature with no electricity; it's purely passive cooling."

Radiative cooling systems could, for example, reduce the electricity required for air conditioning by 10 percent to 15 percent, he said. Such systems at night could also generate enough electricity for LED lighting in homes, which would be a significant development for the billion humans without electricity.

 

Other Stanford faculty research presented includes,

  • Professor Yi Cui, discussed new horizons for energy and climate research as part of a panel. To Cui, the big issue is energy storage to enable greater use of intermittent solar and wind power.
  • Professor Reihold Dauskardt's Spray-on Solar cells
  • Professor Arun Majumdar discussed gigaton-scale solutions for getting to zero greenhouse gas emissions globally from human activity.

 

Excerpted from Precourt Institute "Stanford at CERAWeek: energy storage, net-zero GHG, radiative cooling and perovskite solar cells"

 

Related News

Pages

Subscribe to RSS - research