Designing AI Systems with Steerable Long-Term Dynamics
The feedback that users provide through their choices (e.g. clicks, purchases) is one of the most common types of data readily available for training autonomous systems, and it is widely used in online platforms. However,…
AutoRL Research
This repository contains code for a series of research projects on Automated Reinforcement Learning (AutoRL).
SUMEval 2022
Massively Multilingual Language Models (MMLMs) are trained on around 100 languages of the world, however, most existing multilingual NLP benchmarks provide evaluation data in only a handful of these languages. The languages present in evaluation…
Trusted AI-assisted Programming
Machine learning, in particular Large Language Models, has shown great promise at automating several aspects of programming and software development such as coding, testing, integration, static analysis, verification etc. in recent years. In this project,…
Research Focus: Week of November 17, 2022
Welcome to Research Focus, a new series of blog posts that highlights notable publications, events, code/datasets, new hires and other milestones from across the research community at Microsoft. Microsoft is a proud platinum sponsor of…
Damage Assessment Visualizer
The Damage Assessment Visualizer leverages satellite imagery from a disaster region to visualize conditions of building and structures before and after a disaster. It includes a visual layer on top of the satellite images which…
How to Fine-Tune Vision Models with SGD
imodelsX
Scikit-learn friendly library to interpret, and prompt-engineer text datasets using large language models.