Publications by members of

Brain Tumor Segmentation Using Dense Fully Convolutional Neural Network

Mazhar Shaikh, Ganesh Anand, Gagan Acharya, Abhijit Amrutkar, Varghese Alex, Ganapathy Krishnamurthi

Manual segmentation of brain tumor is often time consuming and the performance of the segmentation varies based on the operators experience. This leads to the requisition of a fully automatic method for brain tumor segmentation. In this paper, we propose the usage of the 100 layer Tiramisu …

International MICCAI Brainlesion Workshop 2018

Towards Incremental Cylindrical Algebraic Decomposition in Maple

Alexander Imani Cowen-Rivers, Matthew England

Cylindrical Algebraic Decomposition (CAD) is an important tool within computational real algebraic geometry, capable of solving many problems for polynomial systems over the reals. It has long been studied by the Symbolic Computation community and has found recent interest in the Satisfiability …

FLOC 2018

RL: Generic reinforcement learning codebase in TensorFlow

Bryan M. Li, Alexander Cowen-Rivers, Piotr Kozakowski, David Tao, Siddhartha Rao Kamalakara, Nitarshan Rajkumar, Hariharan Sezhiyan, Sicong Huang, Aidan N. Gomez

Vast reinforcement learning (RL) research groups, such as DeepMind and OpenAI, have their internal (private) reinforcement learning codebases, which enable quick prototyping and comparing of ideas to many state-of-the-art (SOTA) methods. We argue the five fundamental properties of a sophisticated …

Journal of Open Source Software 2019

| source code

Targeted Dropout

Aidan N. Gomez, Ivan Zhang, Kevin Swersky, Yarin Gal, Geoffrey E. Hinton

Neural networks are extremely flexible models due to their large number of parameters, which is beneficial for learning, but also highly redundant. This makes it possible to compress neural networks without having a drastic effect on performance. We introduce targeted dropout, a strategy for post …


| source code

Unsupervised Cipher Cracking Using Discrete GANs

Aidan N. Gomez, Sicong Huang, Ivan Zhang, Bryan M. Li, Muhammad Osama, Lukasz Kaiser

This work details CipherGAN, an architecture inspired by CycleGAN used for inferring the underlying cipher mapping given banks of unpaired ciphertext and plaintext. We demonstrate that CipherGAN is capable of cracking language data enciphered using shift and Vigenere ciphers to a high degree of …

ICLR 2018

| source code

A generic framework for privacy preserving deep learning

Theo Ryffel, Andrew Trask, Morten Dahl, Bobby Wagner, Jason Mancuso, Daniel Rueckert, Jonathan Passerat-Palmbach

We detail a new framework for privacy preserving deep learning and discuss its assets. The framework puts a premium on ownership and secure processing of data and introduces a valuable representation based on chains of commands and tensors. This abstraction allows one to implement complex …

PPML at NeurIPS 2018

A more efficient approach to perform sensitivity analyses in 0D/1D cardiovascular models

Alessandro Melis, Richard H. Clayton, Alberto Marzo

Ageing effect on pulse wave velocity (PWV) in a single arterial bifurcation is studied by means of a 0D distributed vascular model. A Gaussian process emulator is trained on a small set of simulations, and PWV is computed at system inlet. The emulator is used to predict PWV for a distribution of …

Adversarial Neural Pruning

Divyam Madaan, Sung Ju Hwang

It is well known that neural networks are susceptible to adversarial perturbations and are also computationally and memory intensive which makes it difficult to deploy them in real-world applications where security and computation are constrained. In this work, we aim to obtain both robust and …

Attention Is All You Need

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin

The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the …

NIPS 2017

| source code

Autonomous Trail Following

Masoud Hoveidar-Sefid, Michael Jenkin

Following off-road trails is somewhat more complex than following man-made roads. Trails are unstructuredand typically lack standard markers that characterize roadways. Nevertheless, trails can provide an effectiveset of pathways for off-road navigation. Here we approach the problem of trail …

Autonomous trail following using a pre-trained Deep Neural Network

Masoud Hoveidar-Sefid, Michael Jenkin

Trails are unstructured and typically lack standard markers that characterize roadways; nevertheless, trailscan provide an effective set of pathways for off-road navigation. Here we approach the problem of trailfollowing by identifying the deviation of the robot from the heading …

Bayesian sensitivity analysis of a 1D vascular model with Gaussian process emulators

Alessandro Melis, Richard H. Clayton, Alberto Marzo

One‐dimensional models of the cardiovascular system can capture the physics of pulse waves but involve many parameters. Since these may vary among individuals, patient‐specific models are difficult to construct. Sensitivity analysis can be used to rank model parameters by their effect on outputs and …

Depthwise Separable Convolutions for Neural Machine Translation

Lukasz Kaiser, Aidan N. Gomez, Francois Chollet

Depthwise separable convolutions reduce the number of parameters and computation used in convolutional operations while increasing representational efficiency. They have been shown to be successful in image classification models, both in obtaining better models than previously possible for a given …

ICLR 2018

| source code

GaussianProcesses.jl: A Nonparametric Bayes package for the Julia Language

Jamie Fairbrother, Christopher Nemeth, Maxime Rischard, Johanni Brea, Thomas Pinder

Gaussian processes are a class of flexible nonparametric Bayesian tools that are widely used across the sciences, and in industry, to model complex data sources. Key to applying Gaussian process models is the availability of well-developed open source software, which is available in many programming …

Improved biomechanical metrics of cerebral vasospasm identified via sensitivity analysis of a 1D cerebral circulation model

Alessandro Melis, Fernando Silva De Moura, Ignacio Larrabide, Kévin Janot, R. H. Clayton, Ana Paula Narata, Alberto Marzo

Cerebral vasospasm (CVS) is a life-threatening condition that occurs in a large proportion of those affected by subarachnoid haemorrhage and stroke. CVS manifests itself as the progressive narrowing of intracranial arteries. It is usually diagnosed using Doppler ultrasound, which quantifies blood …

Improved diagnosis of cerebral vasospasm through a sensitivity analysis of a 1D cerebral circulation model

Alessandro Melis, Fernando Moura, Ignacio Larrabide, Richard Clayton, Ana Paula Narata, Alberto Marzo

Cerebral vasospasm (CVS) is the progressive narrowing of cerebral arteries following a haemorrhage in the brain. The primary diagnostic method (Transcranial Doppler) quantifies alterations of blood velocities in the affected vessels, but has low sensitivity when the condition affects the peripheral …

Infer Your Enemies and Know Yourself, Learning in Real-Time Bidding with Partially Observable Opponents

Manxing Du, Alexander I. Cowen-Rivers, Ying Wen, Phu Sakulwongtana, Jun Wang, Mats Brorsson, Radu State

Real-time bidding, as one of the most popular mechanisms for selling online ad slots, facilitates advertisers to reach their potential customers. The goal of bidding optimization is to maximize the advertisers' return on investment (ROI) under a certain budget setting. A straightforward solution is …

Learning Embedding Space for Clustering From Deep Representations

Paras Dahal

Clustering is one of the most fundamental unsupervised tasks in machine learning and is elementary in the exploration of high volume data. Recent works propose using deep neural networks for clustering, owing to their ability to learn powerful representations of the data. In this work, we present a …

2018 IEEE International Conference on Big Data

Moving The Sailing Stone

Enas Tarawneh, David Perrett, Jaspal Singh, Robert Codd-Downey, Masoud Hoveidar Sefid, Shreyansh Jain Jeetmal, Michael Jenkin

Following off-road trails is somewhat more complex than following man-made roads. Trails are unstructuredand typically lack standard markers that characterize roadways. Nevertheless, trails can provide an effectiveset of pathways for off-road navigation. Here we approach the problem of trail …

Neural Variational Inference For Estimating Uncertainty in Knowledge Graph Embeddings

Alexander I. Cowen-Rivers, Pasquale Minervini, Tim Rocktaschel, Matko Bosnjak, Sebastian Riedel, Jun Wang

Recent advances in Neural Variational Inference allowed for a renaissance in latent variable models in a variety of domains involving high-dimensional data. While traditional variational methods derive an analytical approximation for the intractable distribution over the latent variables, here we …


| source code

One Model To Learn Them All

Lukasz Kaiser, Aidan N. Gomez, Noam Shazeer, Ashish Vaswani, Niki Parmar, Llion Jones, Jakob Uszkoreit

Deep learning yields great results across many fields, from speech recognition, image classification, to translation. But for each problem, getting a deep model to work well involves research into the architecture and a long period of tuning. We present a single model that yields good results on a …

Tensor2tensor for neural machine translation

Ashish Vaswani, Samy Bengio, Eugene Brevdo, Francois Chollet, Aidan N. Gomez, Stephan Gouws, Llion Jones, Łukasz Kaiser, Nal Kalchbrenner, Niki Parmar, Ryan Sepassi, Noam Shazeer, Jakob Uszkoreit

Tensor2Tensor is a library for deep learning models that is well-suited for neural machine translation and includes the reference implementation of the state-of-the-art Transformer model.

NIPS 2017

| source code

The Reversible Residual Network: Backpropagation Without Storing Activations

Aidan N. Gomez, Mengye Ren, Raquel Urtasun, Roger B. Grosse

Deep residual networks (ResNets) have significantly pushed forward the state-of-the-art on image classification, increasing in performance as networks grow both deeper and wider. However, memory consumption becomes a bottleneck, as one needs to store the activations in order to calculate gradients …

NIPS 2017

| source code

TimbreTron: A WaveNet(CycleGAN(CQT(Audio))) Pipeline for Musical Timbre Transfer

Sicong Huang, Qiyang Li, Cem Anil, Xuchan Bao, Sageev Oore, Roger B. Grosse

In this work, we address the problem of musical timbre transfer, where the goal is to manipulate the timbre of a sound sample from one instrument to match another instrument while preserving other musical content, such as pitch, rhythm, and loudness. In principle, one could apply image-based style …

ICLR 2019

| source code