
-
Bottleneck Transformers for Visual Recognition
We present BoTNet, a conceptually simple yet powerful backbone architect...
read it
-
Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation
Building instance segmentation models that are data-efficient and can ha...
read it
-
D2RL: Deep Dense Architectures in Reinforcement Learning
While improvements in deep learning architectures have played a crucial ...
read it
-
Evaluating Self-Supervised Pretraining Without Using Labels
A common practice in unsupervised representation learning is to use labe...
read it
-
SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning
Model-free deep reinforcement learning (RL) has been successful in a ran...
read it
-
Reinforcement Learning with Augmented Data
Learning from visual observations is a fundamental yet challenging probl...
read it
-
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
We present CURL: Contrastive Unsupervised Representations for Reinforcem...
read it
-
Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design
Flow-based generative models are powerful exact likelihood models with e...
read it
-
Universal Planning Networks
A key challenge in complex visuomotor control is learning abstract repre...
read it