paper-with-me

Papers

Sparse Attentive Backtracking: Long-Range Credit Assignment in Recurrent Networks

2017-11-07 · ICLR 2018 1 · Nan Rosemary Ke, Anirudh Goyal, Olexa Bilaniuk, Jonathan Binas, Laurent Charlin, Chris Pal, Yoshua Bengio

A major drawback of backpropagation through time (BPTT) is the difficulty of learning long-term dependencies, coming from having to propagate credit information backwards through every single step of the forward computation. This makes BPTT both computationally impractical and biologically implausible. For this reason, full backpropagation through time is rarely used on long sequences, and truncated backpropagation through time is used as a heuristic. However, this usually leads to biased estimates of the gradient in which longer term dependencies are ignored. Addressing this issue, we propose an alternative algorithm, Sparse Attentive Backtracking, which might also be related to principles used by brains to learn long-term dependencies. Sparse Attentive Backtracking learns an attention mechanism over the hidden states of the past and selectively backpropagates through paths with high attention weights. This allows the model to learn long term dependencies while only backtracking for a small number of time steps, not just from the recent past but also from attended relevant past states.

📄 PDF Abstract BibTeX arXiv:1711.02326

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Sparse Attentive Backtracking: Temporal CreditAssignment Through Reminding

2018-09-11 · Nan Rosemary Ke, Anirudh Goyal, Olexa Bilaniuk, Jonathan Binas 외

Learning long-term dependencies in extended temporal sequences requires credit assignment to events far back in the past. The most common method for training recurrent neural networks, back-propagation through time (BPTT…

Temporal Sequences

Sparse Attentive Backtracking: Temporal Credit Assignment Through Reminding

2018-12-01 · NeurIPS 2018 12 · Nan Rosemary Ke, Anirudh Goyal Alias Parth Goyal, Olexa Bilaniuk, Jonathan Binas 외

Learning long-term dependencies in extended temporal sequences requires credit assignment to events far back in the past. The most common method for training recurrent neural networks, back-propagation through time (BPTT…

Temporal Sequences

ARM: Advantage Reward Modeling for Long-Horizon Manipulation

2026-04-03 · Yiming Mao, Zixi Yu, Weixin Mao, Yinhao Li 외 arxiv

Long-horizon robotic manipulation remains challenging for reinforcement learning (RL) because sparse rewards provide limited guidance for credit assignment. Practical policy improvement thus relies on richer intermediate…

Reinforcement LearningOffline RL

An Empirical Analysis of Recurrent Learning Algorithms In Neural Lossy Image Compression Systems

2022-01-27 · Ankur Mali, Alexander Ororbia, Daniel Kifer, Lee Giles

Recent advances in deep learning have resulted in image compression algorithms that outperform JPEG and JPEG 2000 on the standard Kodak benchmark. However, they are slow to train (due to backprop-through-time) and, to th…

DecoderImage Compression

Faster Clustering via Non-Backtracking Random Walks

2017-08-26 · Brian Rappaport, Anuththari Gamage, Shuchin Aeron

This paper presents VEC-NBT, a variation on the unsupervised graph clustering technique VEC, which improves upon the performance of the original algorithm significantly for sparse graphs. VEC employs a novel application …

ClusteringGraph Clustering