Pith. sign in

REVIEW 4 cited by

Unbiasing Truncated Backpropagation Through Time

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1705.08209 v1 pith:M5SGWWQ5 submitted 2017-05-23 cs.NE cs.LG

Unbiasing Truncated Backpropagation Through Time

classification cs.NE cs.LG
keywords truncatedbpttartbpbackpropagationcomputationaltimewhilebenefits
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Truncated Backpropagation Through Time (truncated BPTT) is a widespread method for learning recurrent computational graphs. Truncated BPTT keeps the computational benefits of Backpropagation Through Time (BPTT) while relieving the need for a complete backtrack through the whole data sequence at every step. However, truncation favors short-term dependencies: the gradient estimate of truncated BPTT is biased, so that it does not benefit from the convergence guarantees from stochastic gradient theory. We introduce Anticipated Reweighted Truncated Backpropagation (ARTBP), an algorithm that keeps the computational benefits of truncated BPTT, while providing unbiasedness. ARTBP works by using variable truncation lengths together with carefully chosen compensation factors in the backpropagation equation. We check the viability of ARTBP on two tasks. First, a simple synthetic task where careful balancing of temporal dependencies at different scales is needed: truncated BPTT displays unreliable performance, and in worst case scenarios, divergence, while ARTBP converges reliably. Second, on Penn Treebank character-level language modelling, ARTBP slightly outperforms truncated BPTT.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. HRM-Text: Efficient Pretraining Beyond Scaling

    cs.CL 2026-05 unverdicted novelty 6.0

    A 1B-parameter hierarchical recurrent model pretrained on 40B instruction-response tokens achieves 60.7% MMLU and strong results on ARC-C, DROP, GSM8K, and MATH while using 100-900x fewer tokens than standard baselines.

  2. Generative Recursive Reasoning

    cs.AI 2026-05 unverdicted novelty 6.0

    GRAM turns recursive latent reasoning into a generative probabilistic model via stochastic trajectories and amortized variational inference, claiming better performance on structured reasoning tasks than deterministic...

  3. Generative Recursive Reasoning

    cs.AI 2026-05 unverdicted novelty 6.0

    GRAM is a latent-variable generative model that performs recursive reasoning via stochastic trajectories, trained with amortized variational inference to support multi-hypothesis reasoning and unconditional generation.

  4. Sequential Physics-Constrained Neural Operator Forward Modeling for the $\textit{Norne}$ Reservoir System

    cs.LG 2026-05 unverdicted novelty 5.0

    Presents sequential physics-constrained neural operator models for the Norne reservoir with theoretical stability guarantees and empirical accuracy exceeding 0.99 R² for oil production predictions alongside a 10,000x ...