Search Results for author:
Found 7 papers, 6 papers with code
Provable Repair of Deep Neural Networks
This has motivated a large number of techniques for finding unsafe behavior in DNNs.
SyReNN: A Tool for Analyzing Deep Neural Networks
Formally, DNNs are complicated vector-valued functions which come in a variety of sizes and applications.
Analogy-Making as a Core Primitive in the Software Engineering Toolbox
In this paper, we argue that analogy making should be seen as a core primitive in software engineering.
Abstract Neural Networks
We present a framework parameterized by the abstract domain and activation functions used in the DNN that can be used to construct a corresponding ANN.
A Symbolic Neural Network Representation and its Application to Understanding, Verifying, and Patching Networks
Analysis and manipulation of trained neural networks is a challenging and important problem.
Computing Linear Restrictions of Neural Networks
A linear restriction of a function is the same function with its domain restricted to points on a given line.
DeepThin: A Self-Compressing Library for Deep Neural Networks
For DeepSpeech, DeepThin-compressed networks achieve better test loss than all other compression methods, reaching a 28% better result than rank factorization, 27% better than pruning, 20% better than hand-tuned same-size networks, and 12% better than HashedNets.
