# Research

- Research Areas
- Impact Areas

1 Group Results matching all criteria

2 News Results matching all criteria

## CSAIL hosts first-ever TEDxMIT

Speakers — all women — discuss everything from gravitational waves to robot nurses

## MIT hosts workshop on theoretical foundations of deep learning

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.

18 Group Results

#### Research Group

## Applied Computing Group

We design software for high performance computing, develop algorithms for numerical linear algebra, and research random matrix theory and its applications.

#### Research Group

## Commit Group

We focus on finding novel approaches to improve the performance of modern computer systems without unduly increasing the complexity faced by application developers, compiler writers, or computer architects.

#### Research Group

## Complexity Theory Group

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.

#### Research Group

## Computation and Biology

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.

#### Research Group

## Computation Structures Group

Our mission is fostering the creation and development of high-performance, reliable and secure computing systems that are easy to interact with.

#### Research Group

## Computational Connectomics Group

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

#### Research Group

## Cryptography and Information Security Group

We seek to develop techniques for securing tomorrow's global information infrastructure by exploring theoretical foundations, near-term practical applications, and long-range speculative research.

#### Research Group

## Decentralized Information Group

We are investigating decentralized technologies that affect social change.

#### Research Group

## Geometric Data Processing Group

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

#### Research Group

## Haystack Group

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

#### Research Group

## Multicore Algorithmics

We develop techniques for designing, implementing, and reasoning about multiprocessor algorithms, in particular concurrent data structures for multicore machines and the mathematical foundations of the computation models that govern their behavior.

#### Research Group

## Multimodal Understanding Group

Our objective is to build techniques, software, and hardware that enable natural interaction with

computation.

computation.

#### Research Group

## Quantum Information Science Group

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

#### Research Group

## Supertech Research Group

#### Research Group

## Theory of Computation Group

Theory research at CSAIL covers a broad spectrum of topics, including algorithms, complexity theory, cryptography, distributed systems, parallel computing and quantum computing.

#### Research Group

## Theory of Distributed Systems Group

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.

72 Project Results

#### Project

## [Inactive] T-1000: A Thousand-Core Coherent Shared Memory System

(This project is no longer active.) The T-1000, a prototype system of a thousand realistic processors embedded throughout an ensemble of interconnected FPGAs, seeks to demonstrate the scalability of timestamp-based cache coherence protocols on distributed shared memory systems.

#### Project

## A new way of handling all-to-all broadcast

We design a new all-to-all broadcasts scheme with significantly less communication cost using aggregate signatures.

#### Project

## A Simplified and Extensible Cilk Runtime for Research

CilkS is a new runtime system for the Cilk multithreaded programming environment which makes it easy to experiment with new algorithms, data structures, and programming linguistics.

#### Project

## Active Learning of Models for Planning

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.

#### Project

## Algebraic Techniques for Algorithm Design

We work on improving the algorithms for algebraic problems like matrix multiplication, and using these to design algorithms for fundamental non-algebraic problems.

#### Project

## Algorithmic Aspects of Performance Engineering

The project concerns algorithmic solutions for writing fast codes.

#### Project

## An Algorithmic Theory of Brain Networks

We are developing an algorithmic theory for brain networks, based on simple synchronized stochastic graph-based neural network models.

#### Project

## Approximating the diameter of a directed graph

There is a family of approximation algorithms for computing the diameter of an undirected graph that give a time/accuracy trade-off and our goal is to extend these results to directed graphs.

#### Project

## Aspect-Augmented Adversarial Networks for Domain Adaptation

We propose a novel aspect-augmented adversarial network for cross-aspect and cross-domain adaptation tasks. The effectiveness of our approach suggests the potential application of adversarial networks to a broader range of NLP tasks for improved representation learning, such as machine translation and language generation.

#### Project

## Basing Cryptography on Structured Hardness

We aim to base a variety of cryptographic primitives on complexity theoretic assumptions. We focus on the assumption that there exist highly structured problems --- admitting so called "zero-knowledge" protocols --- that are nevertheless hard to compute

#### Project

## Bayesian Optimization for Global Optimization of Expensive Black-box Functions

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.

#### Project

## Better Models for Ride-Sharing

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

#### Project

## BlueDBM: Distributed Flash Storage for Big Data Analytics

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.

#### Project

## Bridging Theory and Practice in Shared-Memory Parallel Algorithm Design

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

#### Project

## Coresets for Machine Learning Algorithms

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

#### Project

## Data Garbling: Computing on Encrypted Data

We are investigating the limits of computing on encrypted data, with a focus on the private outsourcing of computation over sensitive data.

#### Project

## Deep Inverse Planning

Deep inverse planning for learning from high-dimensional demonstrations

#### Project

## Deterministic Algorithms for Robotic Task and Motion Planning

Our goal is to investigate deterministic algorithms for robotic task and motion planning.

#### Project

## Distributed Algorithms for Dynamic and Noisy Platforms

Distributed systems are now everywhere, for example, in wireless communication networks, distributed data-management systems, coordinated robots, transportation systems, and modern multiprocessors.

#### Project

## Distributed Co-prime Sampling Algorithms

To further parallelize co-prime sampling based sparse sensing, we introduce Diophantine Equation in different algebraic structures to build generalized lattice arrays.

With strong relationship to generalized Chinese Remainder Theorem, the geometry properties in the remainder code space, a special lattice space, are explored.

With strong relationship to generalized Chinese Remainder Theorem, the geometry properties in the remainder code space, a special lattice space, are explored.

#### Project

## Distributed Computation in Ant Colonies

We are interested in applying insights from distributed computing theory to understand how ants and other social insects work together to perform complex tasks such as foraging for food, allocating tasks to workers, and choosing high quality nest sites.

#### Project

## Distributed Robot Garden

The robot garden provides an aesthetically pleasing educational platform that can visualize computer science concepts and encourage young students to pursue programming and robotics.

#### Project

## Diversity-inducing Probability Measures

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.

## Suvrit Sra

#### Project

## Driver-Friendly Bilateral Control for Suppressing Traffic Instabilities

Self-driving cars themselves can solve traffic problems even without global control.

50 People Results

## Cenk Baykal

Graduate Student

## Daryl DeFord

Postdoctoral Associate

## Martin Demaine

Robotics Engineer

## Joanne Hanley

Administrative Assistant II

## Jamey Hicks

Research Affiliate

## Siddhartha Jayanti

Graduate Student

## Kenji Kawaguchi

Graduate Student

## William Kuszmaul

Graduate Student

## Ilia Lebedev

Graduate Student

## Lucas Liebenwein

Graduate Student

## Andrea Lincoln

Graduate Student

63 News Results

## How well can computers connect symptoms to diseases?

A new MIT study finds “health knowledge graphs,” which show relationships between symptoms and diseases and are intended to help with clinical diagnosis, can fall short for certain conditions and patient populations. The results also suggest ways to boost their performance.

## Finding the true potential of algorithms

Using mathematical theory, Virginia Williams coaxes algorithms to run faster or proves they’ve hit their maximum speed.

## Autonomous system improves environmental sampling at sea

Robotic boats could more rapidly locate the most valuable sampling spots in uncharted waters.

## Pushy robots learn the fundamentals of object manipulation

Systems “learn” from novel dataset that captures how pushed objects move, to improve their physical interactions with new objects.

## Recovering “lost dimensions” of images and video

Model could recreate video from motion-blurred images and “corner cameras,” may someday retrieve 3D data from 2D medical images.

## Deep learning with point clouds

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

## Better fetal health - by building a map of the placenta

New technique stretches out MRI scans of placentas so they can be more accurately analyzed, and shows the potential of MRI for pregnancy monitoring.

## MIT’s fleet of autonomous boats can now shapeshift

New capabilities allow “roboats” to change configurations to form pop-up bridges, stages, and other structures.

## Artificial intelligence could help data centers run far more efficiently

MIT system “learns” how to optimally allocate workloads across thousands of servers to cut costs, save energy.

## Why did my classifier just mistake a turtle for a rifle?

Two longtime friends explore how computer vision systems go awry.

## Making it easier to program and protect the web

Professor Adam Chlipala builds tools to help programmers more quickly generate optimized, secure code.

## Automated system generates robotic parts for novel tasks

When designing actuators involves too many variables for humans to test by hand, this system can step in.

## CSAIL hosts first-ever TEDxMIT

Speakers — all women — discuss everything from gravitational waves to robot nurses

## Drag-and-drop data analytics

System lets nonspecialists use machine-learning models to make predictions for medical research, sales, and more.

## Automated cryptocode generator is helping secure the web

System automatically writes optimized algorithms to encrypt data in Google Chrome browsers and web applications.

## Autonomous boats can target and latch onto each other

Fleet of “roboats” could collect garbage or self-assemble into floating structures in Amsterdam’s many canals.

## From one brain scan, more information for medical artificial intelligence

System helps machine-learning models glean training information for diagnosing and treating brain conditions Home Node From one brain scan, more information for medical artificial intelligence

## Q&A: Phillip Isola on the art and science of generative models

Image-translation pioneer discusses the past, present, and future of generative adversarial networks, or GANs.

## CSAIL's Daskalakis wins ACM Grace Murray Hopper Award

Constantinos (“Costis”) Daskalakis, an MIT professor and CSAIL principal investigator, has won the 2018 ACM Grace Murray Hopper Award.

## A novel data-compression technique for faster computer programs

Researchers free up more bandwidth by compressing “objects” within the memory hierarchy.

## Advance boosts efficiency of flash storage in data centers

New architecture promises to cut in half the energy and physical space required to store and manage user data.

## Model learns how individual amino acids determine protein function

Technique could improve machine-learning tasks in protein design, drug testing, and other applications.

## “Particle robot” works as a cluster of simple units

Loosely connected disc-shaped “particles” can push and pull one another, moving en masse to transport objects.