6 papers · 1 filter
On Characterizing the Evolution of Embedding Space of Neural Networks using Algebraic Topology
Suryaka Suresh, Bishshoy Das, Vinayak Abrol +1
We study how the topology of feature embedding space changes as it passes through the layers of a well-trained deep neural network (DNN) through Betti numbers. Motivated by existin…
Data Encoding For Healthcare Data Democratisation and Information Leakage Prevention
Anshul Thakur, Tingting Zhu, Vinayak Abrol +3
The lack of data democratization and information leakage from trained models hinder the development and acceptance of robust deep learning-based healthcare solutions. This paper ar…
Coordinate descent on the orthogonal group for recurrent neural network training
Estelle Massart, Vinayak Abrol
We propose to use stochastic Riemannian coordinate descent on the orthogonal group for recurrent neural network training. The algorithm rotates successively two columns of the recu…
Activation function design for deep networks: linearity and effective initialisation
Michael Murray, Vinayak Abrol, Jared Tanner
The activation function deployed in a deep neural network has great influence on the performance of the network at initialisation, which in turn has implications for training. In t…
An Empirical Study of Derivative-Free-Optimization Algorithms for Targeted Black-Box Attacks in Deep Neural Networks
Giuseppe Ughi, Vinayak Abrol, Jared Tanner
We perform a comprehensive study on the performance of derivative free optimization (DFO) algorithms for the generation of targeted black-box adversarial attacks on Deep Neural Net…
A Model-Based Derivative-Free Approach to Black-Box Adversarial Examples: BOBYQA
Giuseppe Ughi, Vinayak Abrol, Jared Tanner
We demonstrate that model-based derivative free optimisation algorithms can generate adversarial targeted misclassification of deep networks using fewer network queries than non-mo…