pith. machine review for the scientific record. sign in

arxiv: 1211.5063 · v2 · submitted 2012-11-21 · 💻 cs.LG

Recognition: unknown

On the difficulty of training Recurrent Neural Networks

Authors on Pith no claims yet
classification 💻 cs.LG
keywords explodinggradientgradientsissuesnetworksneuralproblemsrecurrent
0
0 comments X
read the original abstract

There are two widely known issues with properly training Recurrent Neural Networks, the vanishing and the exploding gradient problems detailed in Bengio et al. (1994). In this paper we attempt to improve the understanding of the underlying issues by exploring these problems from an analytical, a geometric and a dynamical systems perspective. Our analysis is used to justify a simple yet effective solution. We propose a gradient norm clipping strategy to deal with exploding gradients and a soft constraint for the vanishing gradients problem. We validate empirically our hypothesis and proposed solutions in the experimental section.

This paper has not been read by Pith yet.

discussion (0)

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

Forward citations

Cited by 7 Pith papers

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

  1. Coherent-State Propagation: A Computational Framework for Simulating Bosonic Quantum Systems

    quant-ph 2026-04 unverdicted novelty 8.0

    Coherent-state propagation enables quasi-polynomial classical simulation of bosonic circuits with logarithmically many Kerr gates at exponentially small trace-distance error, with polynomial runtime in the weak-nonlin...

  2. Geometry-Induced Long-Range Correlations in Recurrent Neural Network Quantum States

    quant-ph 2026-04 conditional novelty 7.0

    Dilated RNN wave functions induce power-law correlations for the critical 1D transverse-field Ising model and the Cluster state, unlike the exponential decay of conventional RNN ansatze.

  3. PaLM: Scaling Language Modeling with Pathways

    cs.CL 2022-04 accept novelty 6.0

    PaLM 540B demonstrates continued scaling benefits by setting new few-shot SOTA results on hundreds of benchmarks and outperforming humans on BIG-bench.

  4. Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation

    cs.CL 2016-09 accept novelty 6.0

    GNMT deploys 8-layer LSTMs with attention, wordpieces, low-precision inference, and coverage-penalized beam search to match state-of-the-art on WMT'14 En-Fr and En-De while cutting translation errors by 60% in human e...

  5. Inferring identified hadron production in $pp$ collisions with physics-informed machine learning at the LHC

    hep-ph 2026-05 unverdicted novelty 5.0

    A physics-informed neural network infers pT spectra of pi, K, p, Lambda, and Ks in unmeasured rapidity regions from PYTHIA8 pp collisions at 13.6 TeV, achieving 1.5-5.83% yield uncertainties while reproducing yield ra...

  6. A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidence

    cs.LG 2026-04 unverdicted novelty 4.0

    A conditional Wasserstein GAN generates plausible future SWI drought trajectories for French insurance risk management under climate change.

  7. Preventing overfitting in deep learning using differential privacy

    cs.LG 2026-03 unverdicted novelty 4.0

    Differential privacy techniques can help prevent overfitting and improve generalization in deep neural networks.