How artificial intelligence can help combat systemic racism
MLK Visiting Professor S. Craig Watkins looks beyond algorithm bias to an AI future where models more effectively deal with systemic inequality.
When it comes to AI, can we ditch the datasets?
A machine-learning model for image classification that’s trained using synthetic data can rival one trained on the real thing, a study shows.
Transforming the travel experience for the Hong Kong airport
Students propose solutions to re-imagine the customer experience for Hong Kong’s airport city development.
An “oracle” for predicting the evolution of gene regulation
Researchers create a mathematical framework to examine the genome and detect signatures of natural selection, deciphering the evolutionary past and future of non-coding DNA.
Computational modeling guides development of new materials
Chemical engineers use neural networks to discover the properties of metal-organic frameworks, for catalysis and other applications.
MIT ReACT welcomes first Afghan cohort to its largest-yet certificate program
Empowering a global community of learners in displacement.
MIT Center for Real Estate launches the Asia Real Estate Initiative
New effort empowers MIT researchers to shape real estate’s future and build responsibly and sustainably.
Unlocking new doors to artificial intelligence
MEng graduate students engage with IBM to develop their research skills and solutions to real-world problems.
3 Questions: Fotini Christia on racial equity and data science
A new MIT-wide effort launched by the Institute for Data, Systems, and Society uses social science and computation to address systemic racism.
Injecting fairness into machine-learning models
A new technique boosts models’ ability to reduce bias, even if the dataset used to train the model is unbalanced.
Using artificial intelligence to find anomalies hiding in massive datasets
A new machine-learning technique could pinpoint potential power grid failures or cascading traffic bottlenecks in real time.
Deep-learning technique predicts clinical treatment outcomes
A new methodology simulates counterfactual, time-varying, and dynamic treatment strategies, allowing doctors to choose the best course of action.
Can machine-learning models overcome biased datasets?
A model’s ability to generalize is influenced by both the diversity of the data and the way the model is trained, researchers report.
Research aims to mitigate chemical and biological airborne threats
Lincoln Laboratory leads a large-scale measurement campaign in New York City to improve air dispersion models and emergency protocols.
3 Questions: What a single car can say about traffic
Measuring traffic properties requires vast amounts of data. Meshkat Botshekan, a PhD student working with the MIT CSHub, is discovering a more efficient and affordable physics-inspired alternative.