9 papers
System-Specific Interpreters Make Megasystems Friendlier
Matthew Sotoudeh
Modern operating systems, browsers, and office suites have become megasystems built on millions of lines of code. Their sheer size can intimidate even experienced users and program…
Provable Repair of Deep Neural Networks
Matthew Sotoudeh, Aditya V. Thakur
Deep Neural Networks (DNNs) have grown in popularity over the past decade and are now being used in safety-critical domains such as aircraft collision avoidance. This has motivated…
SyReNN: A Tool for Analyzing Deep Neural Networks
Matthew Sotoudeh, Aditya V. Thakur
Deep Neural Networks (DNNs) are rapidly gaining popularity in a variety of important domains. Formally, DNNs are complicated vector-valued functions which come in a variety of size…
Analogy-Making as a Core Primitive in the Software Engineering Toolbox
Matthew Sotoudeh, Aditya V. Thakur
An analogy is an identification of structural similarities and correspondences between two objects. Computational models of analogy making have been studied extensively in the fiel…
Abstract Neural Networks
Matthew Sotoudeh, Aditya V. Thakur
Deep Neural Networks (DNNs) are rapidly being applied to safety-critical domains such as drone and airplane control, motivating techniques for verifying the safety of their behavio…
A Symbolic Neural Network Representation and its Application to Understanding, Verifying, and Patching Networks
Matthew Sotoudeh, Aditya V. Thakur
Analysis and manipulation of trained neural networks is a challenging and important problem. We propose a symbolic representation for piecewise-linear neural networks and discuss i…