Neil C. Rabinowitz

Neil C. Rabinowitz

Auto-generated profile, based on publicly available information

Neil C. Rabinowitz

DeepMind, London, UK

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 March 2017 in Proceedings of the National Academy of Sciences
Verified
Journal Article
Open access
PDF file
JKJames Kirkpatrick
James Kirkpatrick
RPRazvan Pascanu
Razvan Pascanu
NRNeil Rabinowitz
Neil Rabinowitz
JVJoel Veness
Joel Veness
GDGuillaume Desjardins
Guillaume Desjardins
RHRaia Hadsell
Raia Hadsell
Significance: Deep neural networks are currently the most successful machine-learning technique for solving a variety of tasks, including language translation, image classification, and image generation. One weakness of such models is that, unlike humans, they are unable to learn multiple tasks sequentially. In this work we propose a practical solution to train such models sequentially by protecting the weights important for previous tasks. This approach, inspired by synaptic consolidation in neuroscience, enables state of the art results on multiple reinforcement learning problems experienced sequentially.

9 May 2018 in Nature
Verified
Journal Article
Closed access
PDF file
ABAndrea Banino
Andrea Banino
CBCaswell Barry
Caswell Barry
BUBenigno Uria
Benigno Uria
CBCharles Blundell
Charles Blundell
TLTimothy Lillicrap
Timothy Lillicrap
DKDharshan Kumaran
Dharshan Kumaran

31 May 2019 in Science
Verified
Journal Article
Open access
PDF file
MJMax Jaderberg
Max Jaderberg
WCWojciech M. Czarnecki
Wojciech M. Czarnecki
IDIain Dunning
Iain Dunning
LMLuke Marris
Luke Marris
GLGuy Lever
Guy Lever
TGThore Graepel
Thore Graepel
Artificial teamwork: Artificially intelligent agents are getting better and better at two-player games, but most real-world endeavors require teamwork. Jaderberg et al. designed a computer program that excels at playing the video game Quake III Arena in Capture the Flag mode, where two multiplayer teams compete in capturing the flags of the opposing team. The agents were trained by playing thousands of games, gradually learning successful strategies not unlike those favored by their human counterparts. Computer agents competed successfully against humans even when their reaction times were slowed to match those of humans. Science , this issue p. 859

15 June 2018 in Science
Verified
Journal Article
Closed access
PDF file
SES. M. Ali Eslami
S. M. Ali Eslami
DRDanilo Jimenez Rezende
Danilo Jimenez Rezende
FBFrederic Besse
Frederic Besse
FVFabio Viola
Fabio Viola
AMAri S. Morcos
Ari S. Morcos
DHDemis Hassabis
Demis Hassabis
A scene-internalizing computer program: To train a computer to “recognize” elements of a scene supplied by its visual sensors, computer scientists typically use millions of images painstakingly labeled by humans. Eslami et al. developed an artificial vision system, dubbed the Generative Query Network (GQN), that has no need for such labeled data. Instead, the GQN first uses images taken from different viewpoints and creates an abstract description of the scene, learning its essentials. Next, on the basis of this representation, the network predicts what the scene would look like from a new, arbitrary viewpoint. Science , this issue p. 1204

1 June 2011 in Neuron
Verified
Journal Article
Open access
NRNeil C. Rabinowitz
Neil C. Rabinowitz
BWBen D.B. Willmore
Ben D.B. Willmore
JSJan W.H. Schnupp
Jan W.H. Schnupp
AKAndrew J. King
Andrew J. King
The auditory system must represent sounds with a wide range of statistical properties. One important property is the spectrotemporal contrast in the acoustic environment: the variation in sound pressure in each frequency band, relative to the mean pressure. We show that neurons in ferret auditory cortex rescale their gain to partially compensate for the spectrotemporal contrast of recent stimulation. When contrast is low, neurons increase their gain, becoming more sensitive to small changes in the stimulus, although the effectiveness of contrast gain control is reduced at low mean levels. Gain is primarily determined by contrast near each neuron's preferred frequency, but there is also a contribution from contrast in more distant frequency bands. Neural responses are modulated by contrast over timescales of ∼100 ms. By using contrast gain control to expand or compress the representation of its inputs, the auditory system may be seeking an efficient coding of natural sounds.

2 November 2015 in eLife
Verified
Journal Article
Open access
NRNeil C Rabinowitz
Neil C Rabinowitz
RGRobbe L Goris
Robbe L Goris
MCMarlene Cohen
Marlene Cohen
ESEero P Simoncelli
Eero P Simoncelli
Responses of sensory neurons represent stimulus information, but are also influenced by internal state. For example, when monkeys direct their attention to a visual stimulus, the response gain of specific subsets of neurons in visual cortex changes. Here, we develop a functional model of population activity to investigate the structure of this effect. We fit the model to the spiking activity of bilateral neural populations in area V4, recorded while the animal performed a stimulus discrimination task under spatial attention. The model reveals four separate time-varying shared modulatory signals, the dominant two of which each target task-relevant neurons in one hemisphere. In attention-directed conditions, the associated shared modulatory signal decreases in variance. This finding provides an interpretable and parsimonious explanation for previous observations that attention reduces variability and noise correlations of sensory neurons. Finally, the recovered modulatory signals reflect previous reward, and are predictive of subsequent choice behavior.