Razvan Pascanu

28

Papers

4,643

Total Citations

1

Affiliations

Affiliations

Google DeepMind

Papers (28)

A simple neural network module for relational reasoning

NeurIPS 2017arXiv

Interaction Networks for Learning about Objects, Relations and Physics

NeurIPS 2016arXiv

Distral: Robust multitask reinforcement learning

NeurIPS 2017arXiv

Imagination-Augmented Agents for Deep Reinforcement Learning

NeurIPS 2017arXiv

Sobolev Training for Neural Networks

NeurIPS 2017arXiv

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks

Meta-learning how to Share Credit among Macro-Actions

NeurIPS 2025arXiv

Visual Interaction Networks: Learning a Physics Simulator from Video

Improving fine-grained understanding in image-text pre-training

Universality of Linear Recurrences Followed by Non-linear Projections: Finite-Width Guarantees and Benefits of Complex Eigenvalues

Fine-tuning Reinforcement Learning Models is Secretly a Forgetting Mitigation Problem

Natural Neural Networks

The Tunnel Effect: Building Data Representations in Deep Neural Networks

Sharp Minima Can Generalize For Deep Nets

Mix & Match Agent Curricula for Reinforcement Learning

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

Progress & Compress: A scalable framework for continual learning

Relational recurrent neural networks

Continual Unsupervised Representation Learning

Pointer Graph Networks

Understanding the Role of Training Regimes in Continual Learning

Top-KAST: Top-K Always Sparse Training

Continual World: A Robotic Benchmark For Continual Reinforcement Learning

Powerpropagation: A sparsity inducing weight reparameterisation

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

Disentangling Transfer in Continual Reinforcement Learning

Deep Reinforcement Learning with Plasticity Injection

Learning to Modulate pre-trained Models in RL