paper-with-me

홈 › Papers

Adaptively Truncating Backpropagation Through Time to Control Gradient Bias

2019-05-17 · Christopher Aicher, Nicholas J. Foti, Emily B. Fox

Truncated backpropagation through time (TBPTT) is a popular method for learning in recurrent neural networks (RNNs) that saves computation and memory at the cost of bias by truncating backpropagation after a fixed number of lags. In practice, choosing the optimal truncation length is difficult: TBPTT will not converge if the truncation length is too small, or will converge slowly if it is too large. We propose an adaptive TBPTT scheme that converts the problem from choosing a temporal lag to one of choosing a tolerable amount of gradient bias. For many realistic RNNs, the TBPTT gradients decay geometrically in expectation for large lags; under this condition, we can control the bias by varying the truncation length adaptively. For RNNs with smooth activation functions, we prove that this bias controls the convergence rate of SGD with biased gradients for our non-convex loss. Using this theory, we develop a practical method for adaptively estimating the truncation length during training. We evaluate our adaptive TBPTT method on synthetic data and language modeling tasks and find that our adaptive TBPTT ameliorates the computational pitfalls of fixed TBPTT.

📄 PDF Abstract BibTeX arXiv:1905.07473

Code (1)

aicherc/adaptive_tbptt 공식 구현 pytorch

Tasks

Language ModelingLanguage Modelling

Methods 이 논문이 사용한 방법론

SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

Exploring the Promise and Limits of Real-Time Recurrent Learning

2023-05-30 · Kazuki Irie, Anand Gopalakrishnan, Jürgen Schmidhuber

Real-time recurrent learning (RTRL) for sequence-processing recurrent neural networks (RNNs) offers certain conceptual advantages over backpropagation through time (BPTT). RTRL requires neither caching past activations n…

Diagnostic

On Adaptive Propensity Score Truncation in Causal Inference

2017-07-18 · Cheng Ju, Joshua Schwab, Mark J. Van Der Laan

The positivity assumption, or the experimental treatment assignment (ETA) assumption, is important for identifiability in causal inference. Even if the positivity assumption holds, practical violations of this assumption…

Causal Inference

Neuromorphic Reinforcement Learning for Quadruped Locomotion Control on Uneven Terrain

2026-05-10 · Zhuangyu Han, Abhronil Sengupta arxiv

Reinforcement learning (RL) has enabled robust quadruped locomotion over complex terrain, but most learned controllers are trained offline with backpropagation in massively parallel simulation and deployed as fixed polic…

Reinforcement Learning

Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency

2025-10-22 · Renzhao Liang, Sizhe Xu, Chenggang Xie, Jingru Chen 외 arxiv

Time series forecasting plays a pivotal role in critical domains such as energy management and financial markets. Although deep learning-based approaches (e.g., MLP, RNN, Transformer) have achieved remarkable progress, t…

Time Series ForecastingTime Series Prediction

ABPT: Amended Backpropagation through Time with Partially Differentiable Rewards

2025-01-24 · Fanxing Li, Fangyu Sun, Tianbao Zhang, Danping Zou

Quadrotor control policies can be trained with high performance using the exact gradients of the rewards to directly optimize policy parameters via backpropagation-through-time (BPTT). However, designing a fully differen…