drinkingcoder.github.io
LIFE: Lighting Invariant Flow Estimation
1 min read · Jun 24th · We tackle the problem of estimating flow between two images with large lighting variations. Recent learning-based flow estimation frameworks have shown remarkable performance on image pairs with small…
openreview.net
Disagreement-Regularized Imitation Learning
Jun 2nd · Method for addressing covariate shift in imitation learning using ensemble uncertainty
sites.google.com
Home
4 min read · Jun 2nd · Decision Transformer: Reinforcement Learning via Sequence Modeling Lili Chen*, Kevin Lu*, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Michael Laskin, Pieter Abbeel, Aravind Srinivas*, Igor Mordatch*…
AI Papers
Asymmetric self-play for automatic goal discovery in robotic manipulation
1 min read · May 3rd · We train a single, goal-conditioned policy that can solve many robotic manipulation tasks, including tasks with previously unseen goals and objects. We rely on asymmetric self-play for goal discovery,…
robotics-self-play.github.io
Robotics Asymmetric Self-Play
1 min read · May 3rd · We use asymmetric self-play to train a goal-conditioned policy for complex object manipulation tasks, and the learned policy can zero-shot generalize to many manually designed holdout tasks.
AI Papers
Universal Off-Policy Evaluation
1 min read · May 1st · When faced with sequential decision-making problems, it is often useful to be able to predict what would happen if decisions were made using a new policy. Those predictions must often be based on data…
AI Papers
GOCor: Bringing Globally Optimized Correspondence Volumes into Your Neural Network
1 min read · Apr 22nd · The feature correlation layer serves as a key neural network module in numerous computer vision problems that involve dense correspondences between image pairs. It predicts a correspondence volume by…
monkeyoverflow.com
Thu Nguyen-Phuoc
From 2019 · Personal website for Thu Nguyen-Phuoc, Ph.D. Student in Visual Computing at the Centre for Digital Entertainment, University of Bath.
AI Papers
DualPoseNet: Category-level 6D Object Pose and Size Estimation using Dual Pose Network with Refined Learning of Pose Consistency
1 min read · Mar 14th · Category-level 6D object pose and size estimation is to predict 9 degrees-of-freedom (9DoF) pose configurations of rotation, translation, and size for object instances observed in single, arbitrary…
AI Papers
RGB Matters: Learning 7-DoF Grasp Poses on Monocular RGBD Images
1 min read · Mar 7th · General object grasping is an important yet unsolved problem in the field of robotics. Most of the current methods either generate grasp poses with few DoF that fail to cover most of the success…
agarwl.github.io
Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learning
1 min read · Mar 6th · Reinforcement learning methods trained on few environments rarely learn policies that generalize to unseen environments. To improve generalization, we incorporate the inherent sequential structure in…
AI Papers
Provably Breaking the Quadratic Error Compounding Barrier in Imitation Learning, Optimally
1 min read · Mar 4th · We study the statistical limits of Imitation Learning (IL) in episodic Markov Decision Processes (MDPs) with a state space $\mathcal{S}$. We focus on the known-transition setting where the learner is…
robots.ox.ac.uk
Self-supervised Learning of Interpretable Keypoints from Unlabelled Videos
Feb 27th · Tomas Jakab, Ankush Gupta, Hakan Bilen, Andrea Vedaldi In CVPR, 2020 (Oral presentation) Our method learns to estimate interpretable landmarks without any labelled images. It automatically learns from…
