Code quest

As a grad student, Robert Gallager, SM ’57, ScD ’60, met one of information theory’s central challenges. But nobody realized it for 30 years.

Larry Hardesty, MIT News Office

September 14, 2010

Photo: Mark Ostow

tktkt

More MIT News

Zach Hartwig stands in the Plasma Fusion and Science Center building.

Transforming fusion from a scientific curiosity into a powerful clean energy source

Driven to solve hard problems, Associate Professor Zachary Hartwig is advancing a new approach to commercial fusion energy.

Digital hand and scales in front of electronic iconography

Researchers reduce bias in AI models while preserving or improving accuracy

A new technique identifies and removes the training examples that contribute most to a machine-learning model’s failures.

Illustration of roadways packed with traffic inside of a cell. One protein is asleep at the wheel of a car while another honks behind them. Large trucks labelled ROS fill up the roads. A biker zips past on one side.

Cellular traffic congestion in chronic diseases suggests new therapeutic targets

Chronic diseases like diabetes are prevalent, costly, and challenging to treat. A common denominator driving them may be a promising new therapeutic target.

Skeletal formula of dopamine molecule on a black background

Revisiting reinforcement learning

A detailed new look at dopamine signaling suggests neuroscientists’ model of reinforcement learning may need to be revised.

A bar graph shows left-leaning political bias in four popular AI training datasets.

Study: Some language reward models exhibit political bias

Research from the MIT Center for Constructive Communication finds this effect occurs even when reward models are trained on factual data.

A robot explains a complex graph on a chalkboard and says, “…and a second floor increases the predicted house price by $16,000.”

Enabling AI to explain its predictions in plain language

Using LLMs to convert machine-learning explanations into readable narratives could help users make better decisions about when to trust a model.

MIT News | Massachusetts Institute of Technology

Browse By

Topics

Departments

Centers, Labs, & Programs

Schools

Code quest

More MIT News

Transforming fusion from a scientific curiosity into a powerful clean energy source

Researchers reduce bias in AI models while preserving or improving accuracy

Cellular traffic congestion in chronic diseases suggests new therapeutic targets

Revisiting reinforcement learning

Study: Some language reward models exhibit political bias

Enabling AI to explain its predictions in plain language

Browse By

Topics

Departments

Centers, Labs, & Programs

Schools

Breadcrumb

Code quest

Share this news article on:

More MIT News