#### Research Group

## Algorithms Group

We devise new mathematical tools to tackle the increasing difficulty and importance of problems we pose to computers.

- Impact Areas
- Research Areas

21 Group Results

We devise new mathematical tools to tackle the increasing difficulty and importance of problems we pose to computers.

Our vision is data-driven machine learning systems that advance the quality of healthcare, the understanding of cyber arms races and the delivery of online education.

The MIT Center for Deployable Machine Learning (CDML) works towards creating AI systems that are robust, reliable and safe for real-world deployment.

Our interests span quantum complexity theory, barriers to solving P versus NP, theoretical computer science with a focus on probabilistically checkable proofs (PCP), pseudo-randomness, coding theory, and algorithms.

Our lab focuses on designing algorithms to gain biological insights from advances in automated data collection and the subsequent large data sets drawn from them.

Our group’s goal is to create, based on such microscopic connectivity and functional data, new mathematical models explaining how neural tissue computes.

This community is interested in understanding and affecting the interaction between computing systems and society through engineering, computer science and public policy research, education, and public engagement.

We are investigating decentralized technologies that affect social change.

We aim to develop the science of autonomy toward a future with robots and AI systems integrated into everyday life, supporting people with cognitive and physical tasks.

Our group studies geometric problems in computer graphics, computer vision, machine learning, optimization, and other disciplines.

We are an interdisciplinary group of researchers blending approaches from human-computer interaction, social computing, databases, information management, and databases.

Our mission is to work with policy makers and cybersecurity technologists to increase the trustworthiness and effectiveness of interconnected digital systems.

Our research interests center around the capabilities and limits of quantum computers, and computational complexity theory more generally.

We develop scalable algorithms for analysis of complex, high-dimensional data.

Our goal is to find better ways to make software, and ways to make software better.

We investigate the technologies that support scalable high-performance computing, including hardware, software, and theory.

The Systems CoR is focused on building and investigating large-scale software systems that power modern computers, phones, data centers, and networks, including operating systems, the Internet, wireless networks, databases, and other software infrastructure.

The goal of the Theory of Computation CoR is to study the fundamental strengths and limits of computation as well as how these interact with mathematics, computer science, and other disciplines.

We work on a wide range of problems in distributed computing theory. We study algorithms and lower bounds for typical problems that arise in distributed systems---like resource allocation, implementing shared memory abstractions, and reliable communication.

This CoR takes a unified approach to cover the full range of research areas required for success in artificial intelligence, including hardware, foundations, software systems, and applications.

25 Project Results

We aim to develop a systematic framework for robots to build models of the world and to use these to make effective and safe choices of actions to take in complex scenarios.

The project concerns algorithmic solutions for writing fast codes.

Alloy is a language for describing structures and a tool for exploring them. It has been used in a wide range of applications from finding holes in security mechanisms to designing telephone switching networks. Hundreds of projects have used Alloy for design analysis, for verification, for simulation, and as a backend for many other kinds of analysis and synthesis tools, and Alloy is currently being taught in courses worldwide.

We study the fundamentals of Bayesian optimization and develop efficient Bayesian optimization methods for global optimization of expensive black-box functions originated from a range of different applications.

Traffic is not just a nuisance for drivers: It’s also a public health hazard and bad news for the economy.

BlueDBM is an architecture of computer clusters consisting of fast distributed flash storage and in-storage accelerators, which often outperforms larger and more expensive clusters in applications such as graph analytics.

This project aims to design parallel algorithms for shared-memory machines that are efficient both in theory and also in practice.

Our goal is to design novel data compression techniques to accelerate popular machine learning algorithms in Big Data and streaming settings.

Our goal is to develop new applications and algorithms that leverage the skills of distributed crowdworkers, notably for image and video processing applications.

Data scientists universally report that they spend at least 80% of their time finding data sets of interest, accessing them, cleaning them and assembling them into a unified whole.

The conventional wisdom described in all text books for performing database design is never followed in practice.

Wikipedia is one of the most widely accessed encyclopedia sites in the world, including by scientists. Our project aims to investigate just how far Wikipedia’s influence goes in shaping science.

We aim to understand theory and applications of diversity-inducing probabilities (and, more generally, "negative dependence") in machine learning, and develop fast algorithms based on their mathematical properties.

Developing state-of-the-art tools that process 3D surfaces and volumes

Data often has geometric structure which can enable better inference; this project aims to scale up geometry-aware techniques for use in machine learning settings with lots of data, so that this structure may be utilized in practice.

We are designing new parallel algorithms, optimizations, and frameworks for clustering large-scale graph and geometric data.

Linking probability with geometry to improve the theory and practice of machine learning

Gerrymandering is a direct threat to our democracy, undermining founding principles like equal protection under the law and eroding public confidence in elections.

We are developing a general framework that enforces privacy transparently enabling different kinds of machine learning to be developed that are automatically privacy preserving.

We plan to develop a programming abstraction to enable programmers to write efficient parallel programs to process dynamic graphs.

Developing state-of-the-art deep learning algorithms for analyzing and modeling 3D geometry

Starling is a scalable query execution engine built on cloud function services that computes at a fine granularity, helping people more easily match workload demand.

A polyhedral compiler for expressing image processing, DNN, and linear/tensor algebra applications

System piggybacks on the digital currency’s security protocols to thwart hijacked servers.

40 People Results

Graduate Student

Graduate Student

Graduate Student

Postdoctoral Associate

Graduate Student

Director, Project on Technology, Economy & National Security

Graduate Student

Research Affiliate

Graduate Student

Research Affiliate

Graduate Student

Graduate Student

Postdoctoral Associate

11 News Results

Research aims to make it easier for self-driving cars, robotics, and other applications to understand the 3D world.

Last week MIT’s Institute for Foundations of Data Science (MIFODS) held an interdisciplinary workshop aimed at tackling the underlying theory behind deep learning. Led by MIT professor Aleksander Madry, the event focused on a number of research discussions at the intersection of math, statistics, and theoretical computer science.

May 2, 2018 - Sir Tim Berners-Lee of MIT gave a Dertouzos Distinguished Lecture titled "From Utopia to Dystopia in 29 Short Years."

Harini Suresh, a PhD student at MIT CSAIL, studies how to make machine learning algorithms more understandable and less biased.

This week it was announced that MIT professors and CSAIL principal investigators Shafi Goldwasser, Silvio Micali, Ronald Rivest, and former MIT professor Adi Shamir won this year’s BBVA Foundation Frontiers of Knowledge Awards in the Information and Communication Technologies category for their work in cryptography.

Artificial intelligence (AI) in the form of “neural networks” are increasingly used in technologies like self-driving cars to be able to see and recognize objects. Such systems could even help with tasks like identifying explosives in airport security lines.

We live in the age of big data, but most of that data is “sparse.” Imagine, for instance, a massive table that mapped all of Amazon’s customers against all of its products, with a “1” for each product a given customer bought and a “0” otherwise. The table would be mostly zeroes.

Most modern websites store data in databases, and since database queries are relatively slow, most sites also maintain so-called cache servers, which list the results of common queries for faster access. A data center for a major web service such as Google or Facebook might have as many as 1,000 servers dedicated just to caching.

This week the Association for Computer Machinery presented CSAIL principal investigator Daniel Jackson with the 2017 ACM SIGSOFT Outstanding Research Award for his pioneering work in software engineering. (This fall he also received the ACM SIGSOFT Impact Paper Award for his research method for finding bugs in code.)An EECS professor and associate director of CSAIL, Jackson was given the Outstanding Research Award for his “foundational contributions to software modeling, the creation of the modeling language Alloy, and the development of a widely used tool supporting model verification.”

CSAIL researchers recently helped develop "STEALTH," a system that uses artificial intelligence to combat tax evasion by corporations.

When a power company wants to build a new wind farm, it generally hires a consultant to make wind speed measurements at the proposed site for eight to 12 months. Those measurements are correlated with historical data and used to assess the site’s power-generation capacity.This month CSAIL researchers will present a new statistical technique that yields better wind-speed predictions than existing techniques do — even when it uses only three months’ worth of data. That could save power companies time and money, particularly in the evaluation of sites for offshore wind farms, where maintaining measurement stations is particularly costly.