arXiv (All)
open access
The cost of useful natural gradient updates
What information is needed to turn a natural-gradient direction into a useful finite update? Under a population Kullback-Leibler (KL) budget, we call a step useful if it is feasible and loses at most a fraction $\varepsilon$ of the best feasible gain along the direction. We construct a four-state ex
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- The Statistical Benefits of Multiple Responses for Learning from Demonstrations#1118003
- LearnAlign: Data Selection for LLM Reinforcement Learning with Improved Gradient Alignment#173312
- Backpropagation-free spiking neural networks with the forward-forward algorithm.#159891
- Gradient Cybernetics: The Calculus of Recursion - Synthesis Essay IX - Generative Persistence The Margin Defense: Proving Φ = +0.002 Is Invariant Under Recursive Load#716480
- Learning a Convex Patch-Based Synthesis Model via Deep Equilibrium#235361
Record · ID 1118211
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive