In a new paper, Ghassemi, now an assistant professor in MIT's Department of Electrical Science and Engineering , and three collaborators based at the Computer Science and Artificial Intelligence Laboratory, have probed the roots of the disparities that can arise in machine learning, often causing models that perform well overall to falter when it comes to subgroups for which relatively few data have been collected and utilized in the training process.
The main point of their inquiry is to determine the kinds of subpopulation shifts that can occur and to uncover the mechanisms behind them so that, ultimately, more equitable models can be developed. Given the data available to it, the machine could reach an erroneous conclusion—namely that cows can only be found on grass, not on sand, with the opposite being true for camels. Such a finding would be incorrect, however, giving rise to a spurious correlation, which, Yang explains, is a"special case" among subpopulation shifts—"one in which you have a bias in both the class and the attribute.
A relatively straightforward case would involve just two attributes: the people getting X-rayed are either female or male. If, in this particular dataset, there were 100 males diagnosed with pneumonia for every one female diagnosed with pneumonia, that could lead to an attribute imbalance, and the model would likely do a better job of correctly detecting pneumonia for a man than for a woman.
Improvements to the"encoder," one of the uppermost layers in the neural network, could reduce the problem of attribute imbalance."However, no matter what we did to the encoder or classifier, we did not see any improvements in terms of attribute generalization," Yang says,"and we don't yet know how to address that."There is also the question of assessing how well your model actually works in terms of evenhandedness among different population groups.
We have summarized this news so that you can read it quickly. If you are interested in the news, you can read the full text here. Read more:
United Kingdom Latest News, United Kingdom Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Study suggests antibodies triggered by COVID-19 mRNA vaccination may depend on prior infection historyResearchers evaluated distinct immune responses to mRNA and vector-based COVID-19 vaccines and natural SARS-CoV-2 infections.
Source: NewsMedical - 🏆 19. / 71 Read more »
Magic: The Gathering returns to fairytales for Wilds of EldraineJody's first computer was a Commodore 64, so he remembers having to use a code wheel to play Pool of Radiance. A former music journalist who interviewed everyone from Giorgio Moroder to Trent Reznor, Jody also co-hosted Australia's first radio show about videogames, Zed Games. He's written for Rock Paper Shotgun, The Big Issue, GamesRadar, Zam, Glixel, Five Out of Ten Magazine, and Playboy.com, whose cheques with the bunny logo made for fun conversations at the bank. Jody's first article for PC Gamer was about the audio of Alien Isolation, published in 2015, and since then he's written about why Silent Hill belongs on PC, why Recettear: An Item Shop's Tale is the best fantasy shopkeeper tycoon game, and how weird Lost Ark can get. Jody edited PC Gamer Indie from 2017 to 2018, and he eventually lived up to his promise to play every Warhammer videogame.
Source: pcgamer - 🏆 38. / 67 Read more »
Handshake drama! Richarlison set to snub Michail Antonio prior to kick-off of next Tottenham vs West Ham game after being mocked by him on a podcastWest Ham striker Michail Antonio says Tottenham forward Richarlison will refuse to shake his hand the next time the two teams meet.
Source: goal - 🏆 59. / 63 Read more »
Several vaccines associated with reduced risk of Alzheimer's disease in adults 65 and olderPrior vaccination against tetanus and diphtheria, with or without pertussis (Tdap/Td); herpes zoster (HZ), better known as shingles; and pneumococcus are all associated with a reduced risk for developing Alzheimer's disease, according to new research from UTHealth Houston.
Source: medical_xpress - 🏆 101. / 51 Read more »
Researchers combine deep learning and physics to fix motion-corrupted MRI scansCompared to other imaging modalities like X-rays or CT scans, MRI scans provide high-quality soft tissue contrast. Unfortunately, MRI is highly sensitive to motion, with even the smallest of movements resulting in image artifacts. These artifacts put patients at risk of misdiagnoses or inappropriate treatment when critical details are obscured from the physician. But researchers at MIT may have developed a deep learning model capable of motion correction in brain MRI.
Source: medical_xpress - 🏆 101. / 51 Read more »