On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,982 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “Actor-Mimic”

Compression and Localization in Reinforcement Learning for ATARI Gamescs.LG

This work compresses reinforcement learning models for Atari games, improving localization.

problem Expensive deep neural networks in reinforcement learning.

method Model compression, global max-pooling, Actor-Mimic, weakly supervised localization.

result Compression reduces model size to 3% of original, enabling object localization.

Reinforcement Learning ATARI Games Model Compression Actor-Mimic Localization

2019-04-20abs ↗pdf ↗

Uses and Abuses of the Cross-Entropy Loss: Case Studies in Modern Deep Learningstat.ML

Researchers propose better probabilistic models for deep learning.

problem Using cross-entropy loss for non-categorical data.

method Introducing continuous-categorical distribution and proposing probabilistic alternatives.

result Potential for outperformance in deep learning models with proper probabilistic treatment.

cross-entropy loss deep learning continuous-categorical probabilistic treatment label smoothing

2020-11-10abs ↗pdf ↗