Razvan Pascanu

发表

Relational inductive biases, deep learning, and graph networks pdf

Razvan Pascanu, H. Francis Song, Jessica B. Hamrick, 2018, ArXiv.

Revisiting Natural Gradient for Deep Networks pdf

Razvan Pascanu, Yoshua Bengio, Yoshua Bengio, 2013, ICLR.

Theano: A Python framework for fast computation of mathematical expressions pdf

John Salvatier, Razvan Pascanu, Sina Honari, 2016, ArXiv.

M L ] 2 0 A ug 2 01 3 Pylearn 2 : a machine learning research library pdf

Razvan Pascanu, Yoshua Bengio, Vincent Dumoulin, 2014 .

Meta-learning of Sequential Strategies pdf

Yee Whye Teh, Razvan Pascanu, Nando de Freitas, 2019, ArXiv.

Relational recurrent neural networks

Razvan Pascanu, Oriol Vinyals, Daan Wierstra, 2018, NeurIPS.

On the difficulty of training recurrent neural networks

Razvan Pascanu, Yoshua Bengio, Tomas Mikolov, 2012, ICML.

Natural Neural Networks

Razvan Pascanu, Karen Simonyan, Koray Kavukcuoglu, 2015, NIPS.

Theano: new features and speed improvements pdf

Razvan Pascanu, Yoshua Bengio, Ian J. Goodfellow, 2012, ArXiv.

Sharp Minima Can Generalize For Deep Nets

Razvan Pascanu, Samy Bengio, Yoshua Bengio, 2017, ICML.

Sharp Minima Can Generalize For Deep Nets Supplementary Material

Yoshua Bengio, Samy Bengio, Razvan Pascanu, 2017 .

Sim-to-Real Robot Learning from Pixels with Progressive Nets pdf

Razvan Pascanu, Raia Hadsell, Nicolas Heess, 2016, CoRL.

Understanding the exploding gradient problem pdf

Razvan Pascanu, Yoshua Bengio, Tomas Mikolov, 2012, ArXiv.

Local minima in training of deep networks pdf

Razvan Pascanu, Wojciech Czarnecki, Grzegorz Swirszcz, 2017, ArXiv.

A neurodynamical model for working memory

Razvan Pascanu, Herbert Jaeger, Razvan Pascanu, 2011, Neural Networks.

Memory-based Parameter Adaptation pdf

Razvan Pascanu, Demis Hassabis, Oriol Vinyals, 2018, ICLR.

Vector-based navigation using grid-like representations in artificial agents

Razvan Pascanu, Raia Hadsell, Fabio Viola, 2018, Nature.

Relational Deep Reinforcement Learning pdf

Razvan Pascanu, Murray Shanahan, Yujia Li, 2018, ArXiv.

Policy Distillation pdf

Razvan Pascanu, Sergio Gomez Colmenarejo, Raia Hadsell, 2015, ICLR.

Information asymmetry in KL-regularized RL pdf

Yee Whye Teh, Razvan Pascanu, Wojciech Czarnecki, 2019, ICLR.

Distral: Robust multitask reinforcement learning

Yee Whye Teh, Razvan Pascanu, Raia Hadsell, 2017, NIPS.

Advances in optimizing recurrent networks

Razvan Pascanu, Yoshua Bengio, Nicolas Boulanger-Lewandowski, 2012, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

Imagination-Augmented Agents for Deep Reinforcement Learning

Razvan Pascanu, Demis Hassabis, David Silver, 2017, NIPS.

On the Number of Linear Regions of Deep Neural Networks

Razvan Pascanu, Yoshua Bengio, Kyunghyun Cho, 2014, NIPS.

Theano: A CPU and GPU Math Compiler in Python

Razvan Pascanu, Yoshua Bengio, Guillaume Desjardins, 2010, SciPy.

Deep Learners Benefit More from Out-of-Distribution Examples

Razvan Pascanu, Yoshua Bengio, Dumitru Erhan, 2011, AISTATS.

Metric-Free Natural Gradient for Joint-Training of Boltzmann Machines pdf

Razvan Pascanu, Yoshua Bengio, Aaron C. Courville, 2013, ICLR.

Learning model-based planning from scratch pdf

Razvan Pascanu, Yujia Li, Oriol Vinyals, 2017, ArXiv.

Scalable and Order-robust Continual Learning with Hierarchically Decomposed Networks.

Neil C. Rabinowitz, J. Veness, Razvan Pascanu, 2019 .

How to Construct Deep Recurrent Neural Networks pdf

Razvan Pascanu, Yoshua Bengio, Kyunghyun Cho, 2013, ICLR.

Adapting Auxiliary Losses Using Gradient Similarity pdf

Razvan Pascanu, Wojciech Czarnecki, Balaji Lakshminarayanan, 2018, ArXiv.

Theano: Deep Learning on GPUs with Python

Razvan Pascanu, Ian Goodfellow, Yoshua Bengio, 2012 .

Identifying and attacking the saddle point problem in high-dimensional non-convex optimization

Surya Ganguli, Razvan Pascanu, Yoshua Bengio, 2014, NIPS.

Progressive Neural Networks pdf

Razvan Pascanu, Raia Hadsell, Koray Kavukcuoglu, 2016, ArXiv.

Discovering objects and their relations from entangled scene representations pdf

Razvan Pascanu, Timothy P. Lillicrap, Adam Santoro, 2017, ICLR.

Overcoming catastrophic forgetting in neural networks

Razvan Pascanu, Raia Hadsell, Joel Veness, 2016, Proceedings of the National Academy of Sciences.

Metacontrol for Adaptive Imagination-Based Optimization pdf

Razvan Pascanu, Jessica B. Hamrick, Oriol Vinyals, 2017, ICLR.

Learning to Navigate in Complex Environments pdf

Razvan Pascanu, Misha Denil, Raia Hadsell, 2016, ICLR.

On the number of response regions of deep feed forward networks with piece-wise linear activations

Razvan Pascanu, Yoshua Bengio, Guido Montufar, 2013, 1312.6098.

On the saddle point problem for non-convex optimization pdf

Surya Ganguli, Razvan Pascanu, Yoshua Bengio, 2014, ArXiv.

Progress & Compress: A scalable framework for continual learning pdf

Yee Whye Teh, Razvan Pascanu, Raia Hadsell, 2018, ICML.

Meta-Learning with Latent Embedding Optimization pdf

Razvan Pascanu, Raia Hadsell, Simon Osindero, 2018, ICLR.

Visual Interaction Networks: Learning a Physics Simulator from Video

Razvan Pascanu, Andrea Tacchetti, Theophane Weber, 2017, NIPS.

Combining modality specific deep neural networks for emotion recognition in video

Razvan Pascanu, Christopher Joseph Pal, Yoshua Bengio, 2013, ICMI '13.

On Recurrent and Deep Neural Networks

Razvan Pascanu, 2015 .

Local minima in training of neural networks

Razvan Pascanu, Wojciech Marian Czarnecki, Grzegorz Swirszcz, 2016, 1611.06310.

Sobolev Training for Neural Networks

Razvan Pascanu, Max Jaderberg, Simon Osindero, 2017, NIPS.

Hyperbolic Attention Networks pdf

Razvan Pascanu, Misha Denil, Nando de Freitas, 2018, ICLR.

Learning Algorithms for the Classification Restricted Boltzmann Machine

Razvan Pascanu, Yoshua Bengio, Michael I. Mandel, 2012, J. Mach. Learn. Res..

Been There, Done That: Meta-Learning with Episodic Recall

Zeb Kurth-Nelson, Razvan Pascanu, Charles Blundell, 2018, ICML.

Autotagging music with conditional restricted Boltzmann machines pdf

Razvan Pascanu, Yoshua Bengio, Michael I. Mandel, 2011, ArXiv.

Learning Deep Generative Models of Graphs pdf

Razvan Pascanu, Yujia Li, Chris Dyer, 2018, ICLR 2018.

Embracing Change: Continual Learning in Deep Neural Networks

Andrei A. Rusu, R. Hadsell, Razvan Pascanu, 2020, Trends in Cognitive Sciences.

Reply to Huszár: The elastic weight consolidation penalty is empirically valid

Andrei A. Rusu, Neil C. Rabinowitz, D. Hassabis, 2018, Proceedings of the National Academy of Sciences.

Malware classification with recurrent networks

Razvan Pascanu, Jack W. Stokes, Hermineh Sanossian, 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

Learning a System-ID Embedding Space for Domain Specialization with Deep Reinforcement Learning

Razvan Pascanu, Peter L. Bartlett, Pieter Abbeel, 2018 .

RECURRENT NEURAL NETWORKS

Matthew W. Hoffman, T. Paine, Caglar Gulcehre, 2019 .

Mix & Match – Agent Curricula for Reinforcement Learning [ Appendix ]

Siddhant M. Jayakumar, Wojciech M. Czarnecki, Max Jaderberg, 2018 .

OCAL MINIMA IN TRAINING OF DEEP NETWORKS

Wojciech M. Czarnecki, Razvan Pascanu, G. Swirszcz, 2017 .

Learned-Norm Pooling for Deep Feedforward and Recurrent Neural Networks

Razvan Pascanu, Yoshua Bengio, Kyunghyun Cho, 2013, ECML/PKDD.

LATENT EMBEDDING OPTIMIZATION

Oriol Vinyals, Simon Osindero, Razvan Pascanu, 2018 .

Pylearn2: a machine learning research library

Ian J. Goodfellow, Yoshua Bengio, J. Bergstra, 2013, ArXiv.

EMORY-BASED P ARAMETER A DAPTATION

Adrià Puigdomènech Badia, Siddhant M. Jayakumar, Jack W. Rae, 2018 .

SAUC-E IUB Team

Razvan Pascanu, A. Birk, S. Schwertfeger, 2006 .

How to construct deep recurrent neural networks: Proceedings of the Second International Conference on Learning Representations (ICLR 2014)

Razvan Pascanu, Yoshua Bengio, Kyunghyun Cho, 2014, International Conference on Learning Representations.

Natural Gradient Revisited

Razvan Pascanu, Yoshua Bengio, Yoshua Bengio, 2013, ICLR.

Top-KAST: Top-K Always Sparse Training

Razvan Pascanu, Simon Osindero, Siddhant M. Jayakumar, 2021, NeurIPS.

On the Role of Optimization in Double Descent: A Least Squares Study pdf

Razvan Pascanu, Ilja Kuzborskij, Csaba Szepesv'ari, 2021, NeurIPS.

Transferring Task Goals via Hierarchical Reinforcement Learning

Yee Whye Teh, Razvan Pascanu, Shaobo Hou, 2018 .

Learned-norm pooling for deep neural networks pdf

Razvan Pascanu, Yoshua Bengio, Kyunghyun Cho, 2013, ArXiv.

The CLRS Algorithmic Reasoning Benchmark

Adrià Puigdomènech Badia, D. Budden, R. Hadsell, 2022, ICML.

Pre-training via Denoising for Molecular Property Prediction

James Martens, Y. Teh, P. Battaglia, 2022, ArXiv.

Human Body Recognition using Reproductive Perception Paradigm

Razvan Pascanu, 2007 .

Pointer Graph Networks pdf

Razvan Pascanu, Oriol Vinyals, Charles Blundell, 2020, NeurIPS.

An Empirical Study of Implicit Regularization in Deep Offline RL

Caglar Gulcehre, Georg Ostrovski, A. Doucet, 2022, Trans. Mach. Learn. Res..

Predicting Unreliable Predictions by Shattering a Neural Network

Yoshua Bengio, A. Vedaldi, Razvan Pascanu, 2021, ArXiv.

Wide Neural Networks Forget Less Catastrophically

Seyed Iman Mirzadeh, Razvan Pascanu, Mehrdad Farajtabar, 2021, ICML.

When Does Re-initialization Work?

Y. Teh, Razvan Pascanu, C. Clopath, 2022, ArXiv.

Imagination-Based Decision Making with Physical Models in Deep Neural Networks

Jessica B. Hamrick, Oriol Vinyals, N. Heess, 2016 .

Pre-training via Denoising for Molecular Property Prediction

James Martens, Y. Teh, P. Battaglia, 2022, ICLR.

Probing Transfer in Deep Reinforcement Learning without Task Engineering

Andrei A. Rusu, R. Hadsell, Razvan Pascanu, 2022, CoLLAs.

Disentangling Transfer in Continual Reinforcement Learning

Razvan Pascanu, Maciej Wołczyk, Piotr Milo's, 2022, NeurIPS.

Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet?

Lars Buesing, C. Blundell, Razvan Pascanu, 2022, ArXiv.

NEVIS'22: A Stream of 100 Tasks Sampled from 30 Years of Computer Vision Research

Diego de Las Casas, Andrei A. Rusu, Marc'Aurelio Ranzato, 2022, ArXiv.

Reasoning-Modulated Representations

R. Hadsell, C. Blundell, Razvan Pascanu, 2021, LoG.

Meta-Learning with Warped Gradient Descent

Andrei A. Rusu, R. Hadsell, Razvan Pascanu, 2019, ICLR.

SemPPL: Predicting pseudo-labels for better contrastive representations

Pierre H. Richemond, Lars Buesing, C. Blundell, 2023, ICLR.

Mix&Match - Agent Curricula for Reinforcement Learning pdf

Yee Whye Teh, Razvan Pascanu, Max Jaderberg, 2018, ICML.

Functional Regularisation for Continual Learning

Yee Whye Teh, Razvan Pascanu, Jonathan Schwarz, 2019 .

Ray Interference: a Source of Plateaus in Deep Reinforcement Learning pdf

Razvan Pascanu, Tom Schaul, Joseph Modayil, 2019, ArXiv.

Model compression via distillation and quantization pdf

Dan Alistarh, Razvan Pascanu, Antonio Polino, 2018, ICLR.

Interaction Networks for Learning about Objects, Relations and Physics

Razvan Pascanu, Koray Kavukcuoglu, Matthew Lai, 2016, NIPS.

Improving the Gating Mechanism of Recurrent Neural Networks

Razvan Pascanu, Matthew W. Hoffman, Albert Gu, 2019, ICML.

Pointer Graph Networks pdf

Razvan Pascanu, Charles Blundell, Lars Buesing, 2020, NeurIPS.

Embracing Change: Continual Learning in Deep Neural Networks

Andrei A. Rusu, R. Hadsell, Razvan Pascanu, 2020, Trends in Cognitive Sciences.

A simple neural network module for relational reasoning

Razvan Pascanu, Timothy P. Lillicrap, Adam Santoro, 2017, NIPS.

Task Agnostic Continual Learning via Meta Learning pdf

Yee Whye Teh, Razvan Pascanu, Xu He, 2019, ArXiv.

Understanding the Role of Training Regimes in Continual Learning pdf

Seyed Iman Mirzadeh, Razvan Pascanu, Mehrdad Farajtabar, 2020, NeurIPS.

Stabilizing Transformers for Reinforcement Learning

Razvan Pascanu, H. Francis Song, Raia Hadsell, 2019, ICML.

Behavior Priors for Efficient Reinforcement Learning pdf

Yee Whye Teh, Razvan Pascanu, Hyeonwoo Noh, 2020, J. Mach. Learn. Res..

Exploiting Hierarchy for Learning and Transfer in KL-regularized RL pdf

Yee Whye Teh, Razvan Pascanu, Hyeonwoo Noh, 2019, ArXiv.

Deep reinforcement learning with relational inductive biases

Razvan Pascanu, Murray Shanahan, Yujia Li, 2018, ICLR.

Powerpropagation: A sparsity inducing weight reparameterisation pdf

Yee Whye Teh, Razvan Pascanu, Peter E. Latham, 2021, NeurIPS.

Distilling Policy Distillation pdf

Razvan Pascanu, Max Jaderberg, Simon Osindero, 2019, AISTATS.

BYOL works even without batch statistics pdf

Razvan Pascanu, Michal Valko, Florian Strub, 2020, ArXiv.

Drawing Multiple Augmentation Samples Per Image During Training Efficiently Decreases Test Error

Razvan Pascanu, Andrew Brock, Samuel L. Smith, 2021, ArXiv.

Resurrecting Recurrent Neural Networks for Long Sequences

Caglar Gulcehre, Razvan Pascanu, Samuel L. Smith, 2023, ICML.

A RAD approach to deep mixture models pdf

Razvan Pascanu, Hugo Larochelle, Jascha Sohl-Dickstein, 2019, DGS@ICLR.

Towards Compute-Optimal Transfer Learning

Marc'Aurelio Ranzato, Razvan Pascanu, Alexandre Galashov, 2023, ArXiv.

Linear Mode Connectivity in Multitask and Continual Learning pdf

Seyed Iman Mirzadeh, Mehrdad Farajtabar, Razvan Pascanu, 2020, ICLR.

Architecture Matters in Continual Learning

Seyed Iman Mirzadeh, Razvan Pascanu, Mehrdad Farajtabar, 2022, ArXiv.

Continual World: A Robotic Benchmark For Continual Reinforcement Learning

Razvan Pascanu, Piotr Milo's, Maciej Wołczyk, 2021, NeurIPS.

Functional Regularisation for Continual Learning using Gaussian Processes pdf

Yee Whye Teh, Razvan Pascanu, Jonathan Schwarz, 2019, ICLR.

Continual Unsupervised Representation Learning

Yee Whye Teh, Razvan Pascanu, Raia Hadsell, 2019, NeurIPS.

Understanding plasticity in neural networks

Will Dabney, Clare Lyle, Razvan Pascanu, 2023, ArXiv.

Multiplicative Interactions and Where to Find Them

Yee Whye Teh, Razvan Pascanu, Simon Osindero, 2020, ICLR.

Understanding plasticity in neural networks

Will Dabney, Clare Lyle, Razvan Pascanu, 2023, arXiv.org.

NEVIS'22: A Stream of 100 Tasks Sampled from 30 Years of Computer Vision Research

Diego de Las Casas, Andrei A. Rusu, Marc'Aurelio Ranzato, 2022, ArXiv.

Task-agnostic Continual Learning with Hybrid Probabilistic Models pdf

Razvan Pascanu, Andrew Gordon Wilson, Dushyant Rao, 2021, ArXiv.

Spectral Normalisation for Deep Reinforcement Learning: an Optimisation Perspective

Razvan Pascanu, Lucian Busoniu, Claudia Clopath, 2021, ICML.

Temporal Difference Uncertainties as a Signal for Exploration

Jane X. Wang, N. Heess, André Barreto, 2020, ArXiv.

Deep Reinforcement Learning with Plasticity Injection

Junhyuk Oh, Georg Ostrovski, Will Dabney, 2023, ArXiv.

Learning to Modulate pre-trained Models in RL

S. Hochreiter, Razvan Pascanu, M. Hofmarcher, 2023, ArXiv.

P RE - TRAINING VIA D ENOISING FOR M OLECULAR P ROPERTY P REDICTION

James Martens, Y. Teh, P. Battaglia, 2023 .

Towards Robust and Efficient Continual Language Learning

Marc'Aurelio Ranzato, Razvan Pascanu, Angeliki Lazaridou, 2023, ArXiv.

Overcoming catastrophic forgetting in neural networks

Andrei A. Rusu, Neil C. Rabinowitz, D. Hassabis, 2016, Proceedings of the National Academy of Sciences.

Kalman Filter for Online Classification of Non-Stationary Data

Michalis K. Titsias, Y. Teh, Razvan Pascanu, 2023, ArXiv.

A study on the plasticity of neural networks pdf

Razvan Pascanu, Wojciech Czarnecki, Claudia Clopath, 2021, ArXiv.

Power Norm Based Lifelong Learning for Paraphrase Generations

Andrei A. Rusu, Neil C. Rabinowitz, J. Veness, 2021 .

The Tunnel Effect: Building Data Representations in Deep Neural Networks

Razvan Pascanu, Ehsan Imani, M. Ostaszewski, 2023, ArXiv.

On the number of inference regions of deep feed forward networks with piece-wise linear activations pdf

Razvan Pascanu, Yoshua Bengio, Guido Montúfar, 2013, ICLR.

Low-pass Recurrent Neural Networks - A memory architecture for longer-term correlation discovery pdf

Razvan Pascanu, Rémi Munos, Will Dabney, 2018, ArXiv.

Regularized Behavior Value Estimation pdf

Razvan Pascanu, Sergio Gomez Colmenarejo, Matthew W. Hoffman, 2021, ArXiv.

Latent Space Representations of Neural Algorithmic Reasoners

Razvan Pascanu, Vladimir V. Mirjani'c, Petar Velivckovi'c University of Cambridge, 2023, ArXiv.

Asynchronous Algorithmic Alignment with Cocycles

Tamara von Glehn, Andrew Dudzik, Razvan Pascanu, 2023, ArXiv.

Under review as a conference paper at ICLR 2017 D ISCOVERING OBJECTS AND THEIR RELATIONS FROM ENTANGLED SCENE REPRESENTATIONS

T. Lillicrap, P. Battaglia, David Raposo, 2016 .

On the Universality of Linear Recurrences Followed by Nonlinear Projections

Caglar Gulcehre, Razvan Pascanu, Samuel L. Smith, 2023, arXiv.org.

Promoting Exploration in Memory-Augmented Adam using Critical Momenta

Reza Babanezhad Harikandeh, Razvan Pascanu, S. Lacoste-Julien, 2023, arXiv.org.

Visual Interaction Networks: Learning a Physics Simulator from Video pdf

Razvan Pascanu, Andrea Tacchetti, Theophane Weber, 2017, NIPS.

Test Sample Accuracy Scales with Training Sample Density in Neural Networks pdf

Razvan Pascanu, Yoshua Bengio, Andrea Vedaldi, 2021, CoLLAs.

Understanding plasticity in neural networks

Will Dabney, Clare Lyle, Razvan Pascanu, 2023, ICML.

When Does Re-initialization Work?

Y. Teh, Razvan Pascanu, C. Clopath, 2022, ICBINB.

LiRo: Benchmark and leaderboard for Romanian language tasks

Radu Tudor Ionescu, Dani Yogatama, Razvan Pascanu, 2021, NeurIPS Datasets and Benchmarks.

Uncovering mesa-optimization algorithms in Transformers

Razvan Pascanu, B. A. Y. Arcas, Max Vladymyrov, 2023, ArXiv.

Deep Self-Taught Learning for Handwritten Character Recognition pdf

Razvan Pascanu, Yoshua Bengio, Dumitru Erhan, 2010, ArXiv.

Relational inductive biases, deep learning, and graph networks

Jessica B. Hamrick, Kelsey R. Allen, George E. Dahl, 2018, ArXiv.

Learning to Modulate pre-trained Models in RL

S. Hochreiter, Razvan Pascanu, M. Hofmarcher, 2023, NeurIPS.

Deep Reinforcement Learning with Plasticity Injection

Junhyuk Oh, Georg Ostrovski, Will Dabney, 2023, ArXiv.

Multi-view learning and deep learning for heterogeneous biological data to maintain oral health

J. Bergstra, Pascal Lamblin, Razvan Pascanu, 2016 .

Continual Learning: Applications and the Road Forward

Tyler L. Hayes, T. Tuytelaars, Razvan Pascanu, 2023, ArXiv.

Been There, Done That: Meta-Learning with Episodic Recall

Siddhant M. Jayakumar, Jane X. Wang, C. Blundell, 2018, ICML.

Asynchronous Algorithmic Alignment with Cocycles

Tamara von Glehn, Andrew Dudzik, Razvan Pascanu, 2023, ArXiv.

Vector-based navigation using grid-like representations in artificial agents

Neil C. Rabinowitz, T. Lillicrap, K. Kavukcuoglu, 2018, Nature.

Combining modality specific deep neural networks for emotion recognition in video

Aaron C. Courville, Samira Ebrahimi Kahou, Raul Chandias Ferrari, 2013, ICMI '13.

Block Mean Approximation for Efficient Second Order Optimization pdf

Razvan Pascanu, Yao Lu, Richard I. Hartley, 2018, ArXiv.

One-Step Diffusion Distillation via Deep Equilibrium Models

Siddhant M. Jayakumar, J. Z. Kolter, Raphael Lopez Kaufman, 2023, NeurIPS.

FAST PROTOTYPING OF AN AUTONOMOUS UNDERWATER VEHICLE WITH THE CUBESYSTEM

Andreas Birk, Razvan Pascanu, Stefan Videv, 2007 .