Biswa Sengupta

is this you? claim profile

Head of Machine Learning at Huawei Technologies since 2017, Visiting Senior Research Scientist at Imperial College London since 2016, Chief Research Scientist at Cortexica Vision Systems from 2016-2017, Senior Research Scientist at University of Cambridge 2016-2017, Research Scientist at University College London from 2011-2016, Research Scientist at Indian Institute of Science from 2011-2012, Research Scientist at LMU - Ludwig-Maximilians-Universität München from 2008-2011, Research Assistant at Hertie-Institute for Clinical Brain Research 2007, Research Assistant at Max Planck Institute for Biological Cybernetics from 2005-2007, Research Scientist at University of York 2004-2005, PhD in Computation and Neural Systems at University of Cambridge 2007-2011

How Robust are Deep Neural Networks?

Convolutional and Recurrent, deep neural networks have been successful i...
04/30/2018 ∙ by Biswa Sengupta, et al. ∙ 0 ∙ share

read it
Conditional Autoencoders with Adversarial Information Factorization

Generative models, such as variational auto-encoders (VAE) and generativ...
11/14/2017 ∙ by Antonia Creswell, et al. ∙ 0 ∙ share

read it
LatentPoison - Adversarial Attacks On The Latent Space

Robustness and security of machine learning (ML) systems are intertwined...
11/08/2017 ∙ by Antonia Creswell, et al. ∙ 0 ∙ share

read it
GeoSeq2Seq: Information Geometric Sequence-to-Sequence Networks

The Fisher information metric is an important foundation of information ...
10/25/2017 ∙ by Alessandro Bay, et al. ∙ 0 ∙ share

read it
Generative Adversarial Networks: An Overview

Generative adversarial networks (GANs) provide a way to learn deep repre...
10/19/2017 ∙ by Antonia Creswell, et al. ∙ 0 ∙ share

read it
StackSeq2Seq: Dual Encoder Seq2Seq Recurrent Networks

A widely studied non-deterministic polynomial time (NP) hard problem lie...
10/11/2017 ∙ by Alessandro Bay, et al. ∙ 0 ∙ share

read it
Sequence stacking using dual encoder Seq2Seq recurrent networks

A widely studied non-polynomial (NP) hard problem lies in finding a rout...
10/11/2017 ∙ by Alessandro Bay, et al. ∙ 0 ∙ share

read it
Approximating meta-heuristics with homotopic recurrent neural networks

Much combinatorial optimisation problems constitute a non-polynomial (NP...
09/07/2017 ∙ by Alessandro Bay, et al. ∙ 0 ∙ share

read it
Pillar Networks++: Distributed non-parametric deep and wide networks

In recent work, it was shown that combining multi-kernel based support v...
08/18/2017 ∙ by Biswa Sengupta, et al. ∙ 0 ∙ share

read it
Multi-kernel learning of deep convolutional features for action recognition

Image understanding using deep convolutional network has reached human-l...
07/21/2017 ∙ by Biswa Sengupta, et al. ∙ 0 ∙ share

read it
Bayesian Belief Updating of Spatiotemporal Seizure Dynamics

Epileptic seizure activity shows complicated dynamics in both space and ...
05/20/2017 ∙ by Gerald K Cooray, et al. ∙ 0 ∙ share

read it
Approximate Bayesian inference as a gauge theory

In a published paper [Sengupta, 2016], we have proposed that the brain (...
05/17/2017 ∙ by Biswa Sengupta, et al. ∙ 0 ∙ share

read it

Biswa Sengupta

Featured Co-authors

Vincent Dumoulin

Kai Arulkumaran

Antonia Creswell

Anil A Bharath

Alessandro Bay

Karl Friston

Yu Qian

Karl J. Friston

Tom White

Louis Lemieux

Gerald K Cooray

How Robust are Deep Neural Networks?

Conditional Autoencoders with Adversarial Information Factorization

LatentPoison - Adversarial Attacks On The Latent Space

GeoSeq2Seq: Information Geometric Sequence-to-Sequence Networks

Generative Adversarial Networks: An Overview

StackSeq2Seq: Dual Encoder Seq2Seq Recurrent Networks

Sequence stacking using dual encoder Seq2Seq recurrent networks

Approximating meta-heuristics with homotopic recurrent neural networks

Pillar Networks++: Distributed non-parametric deep and wide networks

Multi-kernel learning of deep convolutional features for action recognition

Bayesian Belief Updating of Spatiotemporal Seizure Dynamics

Approximate Bayesian inference as a gauge theory