Chrono Gate Initialization Boosts LSTM/GRU Performance on Long Dependencies

作者 PulseAugur 编辑部 · [3 个来源] · 2026-10-21 01:06

一篇新发表在arXiv上的研究，探讨了在长短期记忆（LSTM）和门控循环单元（GRU）网络中，门控初始化技术与循环 dropout 和 zoneout 相比，在学习长依赖方面的有效性。研究人员在合成任务上训练了这些网络，发现 chrono gate initialization 显著提高了性能，使得所有运行在 5000 次更新内都达到了验证标准。相比之下，传统的初始化方法则表现不佳，dropout 通常会延迟学习，而 zoneout 在特定条件下仅提供有限的益处。 AI

影响这项研究为优化需要长期记忆的循环神经网络在序列建模应用中的性能提供了见解。

排序理由该集群包含一篇发表在arXiv上的学术论文，详细介绍了对神经网络架构和训练技术的对照研究。

在 arXiv cs.NE (Neural & Evolutionary) 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。我们如何撰写摘要 →

Chrono Gate Initialization Boosts LSTM/GRU Performance on Long Dependencies

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Research; 该集群包含一篇发表在arXiv上的学术论文，详细介绍了对神经网络架构和训练技术的对照研究。
Source corroboration: 3 independent sources; Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 5 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring: +1 source(s) since last score; New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [3]

arXiv cs.LG TIER_1 English(EN) · Oleksandr Kholodniak, Oleksandr Prokhorov · 2026-10-22 04:00

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks

arXiv:2610.29034v1 Announce Type: new Abstract: Dropout and zoneout are routinely added to long short-term memory (LSTM) and gated recurrent unit (GRU) networks, but, to our knowledge, their placements have not been crossed with gate initialisation in a factorial design, and it i…
arXiv cs.NE (Neural & Evolutionary) TIER_1 English(EN) · Oleksandr Prokhorov · 2026-10-21 01:06

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks

Dropout and zoneout are routinely added to long short-term memory (LSTM) and gated recurrent unit (GRU) networks, but, to our knowledge, their placements have not been crossed with gate initialisation in a factorial design, and it is not known whether early gradient propagation t…
Hugging Face Daily Papers TIER_1 English(EN) · 2026-10-21 01:06

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks

Dropout and zoneout are routinely added to long short-term memory (LSTM) and gated recurrent unit (GRU) networks, but, to our knowledge, their placements have not been crossed with gate initialisation in a factorial design, and it is not known whether early gradient propagation t…

本文如何被排名

报道来源 [3]

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks

Gate Initialisation versus Recurrent Dropout and Zoneout in Learning Long Dependencies: A Controlled Study of LSTM and GRU Networks on Synthetic Tasks

相关实体

相关话题