The latest technology and digital news on the web

Human-centric AI news and analysis

How breakthrough computers can save us from the coronavirus

A team of advisers at Penn State University afresh began exploring activating solutions for drug analysis in hopes of barrier upon the cure for COVID-19. Hundreds of other analysis teams, including those backed by big tech and big pharma, are alive the same problem. But this study is different. The Penn State team is bringing breakthrough AI to the fight.

Traditionally, drug discovery’s been a assiduous action involving trial-and-error. We’re talking eyeballs on microscopes, pen and paper graphs, and bags of audible negatives for every accessible positive. The advent of deep acquirements sped things up a bit.

As Neural’s own Thomas Macaulay afresh acicular out in their article, “Drug analysis might be the best use of AI to tackle the pandemic,” the use of algorithms to sift through near-infinite actinic combinations is a gamechanger. One startup, BenevolentAI, believes its system may have even apparent a abeyant analysis for COVID-19 (its currently in trials).

But what if that analysis doesn’t pan out? How long before the machines find the next abeyant cure? What we need is a method by which we can more bound ascertain actinic combinations that could work as treatments.

Enter the Penn State analysis team led by architect Swaroop Ghosh.

Ghosh’s team hopes to decidedly speed up the action of award a cure or analysis for COVID-19 by acclamation a root botheration with archetypal algorithms for drug discovery, a lack of data. Ghosh told Penn State’s Sarah Small:

High-performance accretion such as supercomputers and bogus intelligence can help advance this action by screening billions of actinic compounds bound to find accordant drug candidates. This access works when enough actinic compounds are accessible in the pipeline, but abominably this is not true for COVID-19. This activity will analyze breakthrough apparatus acquirements to unlock new capabilities in drug analysis by breeding circuitous compounds quickly.

Ghosh is alive with several electrical engineering and computer science acceptance to work on optimizing a breakthrough AI system. This work is exploratory, acceptation there’s no agreement annihilation useful will come out of it. But admitting the fact that breakthrough accretion is in its infancy, there’s hope.

One of the most-mentioned use cases for breakthrough apparatus acquirements algorithms is this sort of discovery. Breakthrough computers excel at these problems because they aren’t accountable by archetypal binary ciphering limitations. With breakthrough computers, advisers can find more avenues to analysis in a beneath period of time using less data.

For more information, check out Ghosh’s work here.

Published May 6, 2020 — 17:25 UTC

Hottest related news