#Bayesian

7 stories · 0 in 7 days · 1 in 30 days

arXiv 1 ATARI 1 DeepSeek-R1 1 ICL 1 LLM 1 QVIRL 1 RLHF 1 SFT 1

Stories, products and papers tagged "Bayesian", newest first.

Oct 26

Papers Builders09:26

Chinese contentBAARC框架：用不确定性驱动拉曼光谱部分测量与材料分类

搞光谱或科学计算的朋友可以看看，这篇把硬件采谱和贝叶斯深度学习做成闭环，只用部分波段就能完成材料识别。

#Bayesian #arXiv

arXiv cs.LG@Adithya Ashok Chalain Valapil et al. (3)Source

Sep 21

Papers BuildersSelected11:16

Bayesian Chronicle Agents Add Controllable Beliefs to LLM Social Simulation

Unlike implicit in-context opinion shifts, BCA lets social-simulation builders verify a stubbornness setting numerically, matching Friedkin–Johnsen fixed points at R² 0.93–0.99 and recovering prescribed parameters perfectly across four LLMs.

#LLM #agents #Bayesian

arXiv cs.AI@Hafsa Akbar et al. (4)Source

Sep 7

Papers BuildersSelected73°09:22

Bayesian Framework Unifies AI Training Methods

The framework explains why few-shot prompting has mixed effects on RL-tuned models and provides insights for modern reasoning pipelines like DeepSeek-R1 and o1-style models.

#Bayesian #RLHF #SFT #ICL

arXiv: DeepSeek@Junxin FanSource

Aug 18

Papers15:34

Chinese contentQVIRL：基于Q值的变分逆强化学习方法

想学逆强化学习？QVIRL能边学边给不确定性，还能主动问你要数据，比老方法稳多了。

#QVIRL #Bayesian #active learning #ATARI

arXiv cs.LG@Ondrej Bajgar et al. (5)Source

Aug 11

Papers10:52

Chinese contentERRLESS：熵正则强化学习实现贝叶斯符号回归

想找能处理噪声数据、还能告诉你不确定性有多大的回归方法？ERRLESS用强化学习从后验里采样表达式，Feynman基准上表现不错，表达式还短。