Deepak Pathak

is this you? claim profile

∙ 1 follower
Carnegie Mellon University ∙ berkeley college

Deepak Pathak is a PhD student focused on computer vision and machine learning.

Planning in Learned Latent Action Spaces for Generalizable Legged Locomotion

Hierarchical learning has been successful at learning generalizable loco...
08/27/2020 ∙ by Tianyu Li, et al. ∙ 0 ∙ share

read it
Learning Long-term Visual Dynamics with Region Proposal Interaction Networks

Learning long-term dynamics models is the key to understanding physical ...
08/05/2020 ∙ by Haozhi Qi, et al. ∙ 2 ∙ share

read it
One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Control

Reinforcement learning is typically concerned with learning control poli...
07/09/2020 ∙ by Wenlong Huang, et al. ∙ 9 ∙ share

read it
Locally Masked Convolution for Autoregressive Models

High-dimensional generative models have many applications including imag...
06/22/2020 ∙ by Ajay Jain, et al. ∙ 60 ∙ share

read it
Planning to Explore via Self-Supervised World Models

Reinforcement learning allows solving complex tasks, however, the learni...
05/12/2020 ∙ by Ramanan Sekar, et al. ∙ 2 ∙ share

read it
Sparse Graphical Memory for Robust Planning

To operate effectively in the real world, artificial agents must act fro...
03/13/2020 ∙ by Michael (Misha) Laskin, et al. ∙ 16 ∙ share

read it
Third-Person Visual Imitation Learning via Decoupled Hierarchical Controller

We study a generalized setup for learning from demonstration to build an...
11/21/2019 ∙ by Pratyusha Sharma, et al. ∙ 0 ∙ share

read it
Self-Supervised Exploration via Disagreement

Efficient exploration is a long-standing problem in sensorimotor learnin...
06/10/2019 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Image Inpainting

Fills holes in images

05/17/2019 Deepak Pathak ∙ 29 ∙ share

try it
Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity

Contemporary sensorimotor learning approaches typically start with an ex...
02/14/2019 ∙ by Deepak Pathak, et al. ∙ 12 ∙ share

read it
Large-Scale Study of Curiosity-Driven Learning

Reinforcement learning algorithms rely on carefully engineering environm...
08/13/2018 ∙ by Yuri Burda, et al. ∙ 0 ∙ share

read it
Compositional GAN: Learning Conditional Image Composition

Generative Adversarial Networks (GANs) can produce images of surprising ...
07/19/2018 ∙ by Samaneh Azadi, et al. ∙ 16 ∙ share

read it
Learning Instance Segmentation by Interaction

We present an approach for building an active agent that learns to segme...
06/21/2018 ∙ by Deepak Pathak, et al. ∙ 2 ∙ share

read it
Zero-Shot Visual Imitation

The current dominant paradigm for imitation learning relies on strong su...
04/23/2018 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Investigating Human Priors for Playing Video Games

What makes humans so good at solving seemingly complex video games? Unli...
02/28/2018 ∙ by Rachit Dubey, et al. ∙ 0 ∙ share

read it
Toward Multimodal Image-to-Image Translation

Many image-to-image translation problems are ambiguous, as a single inpu...
11/30/2017 ∙ by Jun-Yan Zhu, et al. ∙ 0 ∙ share

read it
Curiosity-driven Exploration by Self-supervised Prediction

In many real-world scenarios, rewards extrinsic to the agent are extreme...
05/15/2017 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Learning Features by Watching Objects Move

This paper presents a novel yet intuitive approach to unsupervised featu...
12/19/2016 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Context Encoders: Feature Learning by Inpainting

We present an unsupervised visual feature learning algorithm driven by c...
04/25/2016 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Constrained Convolutional Neural Networks for Weakly Supervised Segmentation

We present an approach to learn a dense pixel-wise labeling from image-l...
06/11/2015 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Fully Convolutional Multi-Class Multiple Instance Learning

Multiple instance learning (MIL) can reduce the need for costly annotati...
12/22/2014 ∙ by Deepak Pathak, et al. ∙ 0 ∙ share

read it
Detector Discovery in the Wild: Joint Multiple Instance and Representation Learning

We develop methods for detector learning which exploit joint training ov...
12/02/2014 ∙ by Judy Hoffman, et al. ∙ 0 ∙ share

read it

Deepak Pathak

Featured Co-authors

Sergey Levine

Pieter Abbeel

Trevor Darrell

Jitendra Malik

Kate Saenko

Abhinav Gupta

Ross Girshick

Alexei A. Efros

Kostas Daniilidis

Xiaolong Wang

Thomas L. Griffiths

Planning in Learned Latent Action Spaces for Generalizable Legged Locomotion

Learning Long-term Visual Dynamics with Region Proposal Interaction Networks

One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Control

Locally Masked Convolution for Autoregressive Models

Planning to Explore via Self-Supervised World Models

Sparse Graphical Memory for Robust Planning

Third-Person Visual Imitation Learning via Decoupled Hierarchical Controller

Self-Supervised Exploration via Disagreement

Image Inpainting

Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity

Large-Scale Study of Curiosity-Driven Learning

Compositional GAN: Learning Conditional Image Composition

Learning Instance Segmentation by Interaction

Zero-Shot Visual Imitation

Investigating Human Priors for Playing Video Games

Toward Multimodal Image-to-Image Translation

Curiosity-driven Exploration by Self-supervised Prediction

Learning Features by Watching Objects Move

Context Encoders: Feature Learning by Inpainting

Constrained Convolutional Neural Networks for Weakly Supervised Segmentation

Fully Convolutional Multi-Class Multiple Instance Learning

Detector Discovery in the Wild: Joint Multiple Instance and Representation Learning