This CoR brings together researchers at CSAIL working across a broad swath of application domains. Within these lie novel and challenging machine learning problems serving science, social science and computer science.
This CoR aims to develop AI technology that synthesizes symbolic reasoning, probabilistic reasoning for dealing with uncertainty in the world, and statistical methods for extracting and exploiting regularities in the world, into an integrated picture of intelligence that is informed by computational insights and by cognitive science.
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.
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 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.
To enable privacy preservation in decentralized optimization, differential privacy is the most commonly used approach. However, under such scenario, the trade-off between accuracy (even efficiency) and privacy is inevitable. In this project, distributed numerical optimization scheme incorporated with lightweight cryptographic information sharing are explored. The affect on the convergence rate from network topology is considered.
Predicting the number of clock cycles a processor takes to execute a block of assembly instructions in steady-state (the throughput) is important for both compiler designers and performance engineers.
However, building an analytical model to do so is especially complicated in modern x86-64 Complex Instruction Set Computer (CISC) machines with sophisticated processor microarchitectures in that it is tedious, error-prone, and must be performed from scratch for each processor generation.
Ithemal is the first tool that learns to predict the throughput of a set of instructions. It does so more accurately than state-of-the-art hand-written tools currently used in compiler backends and static machine code analyzers. In particular, Ithemal has less than half the error of state-of-the-art analytical models (LLVM's llvm-mca and Intel's IACA).
We develop algorithms, systems and software architectures for automating reconstruction of accurate representations of neural tissue structures, such as nanometer-scale neurons' morphology and synaptic connections in the mammalian cortex.
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.
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.
MIT’s Amar Gupta and his wife Poonam were on a trip to Los Angeles in 2016 when she fell and broke both wrists. She was whisked by ambulance to a reputable hospital. But staff informed the couple that they couldn’t treat her there, nor could they find another local hospital that would do so. In the end, the couple was forced to take the hospital’s stunning advice: return to Boston for treatment.