https://andreasloukas.blog/2020/10/31/erdos-goes-neural/ 一年多了才看明白….以概率法的形式解组合优化问题，把概率分布用神经网络来拟合（网络输出的是分布的参数）。亮点是对于这个神经网络的损失函数如何导出的。大概可以用在一些tricky问题上，有兴趣的可以评论里聊聊。
andreas loukas
Erdős goes neural: unsupervised learning of combinatorial algorithms with neural networks
This post summarizes our recent work “Erdős goes neural: an unsupervised learning framework for combinatorial optimization on graphs” (bibtex) presented as an oral contribution at NeurI…
https://thegradient.pub/prompting/
大概是一种更科学和时髦的Pretrained-LM用法：不再fine-tuning而使用prompt
大概是一种更科学和时髦的Pretrained-LM用法：不再fine-tuning而使用prompt
The Gradient
Prompting: Better Ways of Using Language Models for NLP Tasks
A review of recent advances in prompts.
Vintage NLP for Work
https://thegradient.pub/prompting/ 大概是一种更科学和时髦的Pretrained-LM用法：不再fine-tuning而使用prompt
Twitter
Pengfei Liu
What is prompt-based learning, and what challenges are there? Will it be a new paradigm or a way for human-PLMs communication? How does it connect with other research and how to position it in the evolution of the NLP research paradigm? We released a systematic…
Forwarded from 好书分享频道 (神机妙算 文武双全)
内容简介 · · · · · ·
本书即介绍三十年来语义学取得的进展,对各种观点作出 讨论。本书上篇讨论主要有哲学家、逻辑学家,还有语言学 家提出的语义理论,下篇讨论语言学领域内语义研究的重 点课题。
本书即介绍三十年来语义学取得的进展,对各种观点作出 讨论。本书上篇讨论主要有哲学家、逻辑学家,还有语言学 家提出的语义理论,下篇讨论语言学领域内语义研究的重 点课题。
虽然是1995年的书了，但还不错（相对于很多新出版的讲义），补充了很多做CS出身的缺少的语义学知识。至于为什么需要这些基础知识以及为什么有这么多理论，可能就是因为至今semantics是什么也没个共识😂
找到一组挺不错的上古parsing算法简介，这年头transformer力压群雄以后大家又开始怀念grammar带来的inductive bias了，爬上来以后发现黑格尔早已站在AI的高地上（
https://www.borealisai.com/en/blog/tutorial-15-parsing-i-context-free-grammars-and-cyk-algorithm/
CFG和CYK parsing算法
https://www.borealisai.com/en/blog/tutorial-18-parsing-ii-wcfgs-inside-algorithm-and-weighted-parsing/
WCFG和CYK改良的inside算法
https://www.borealisai.com/en/blog/tutorial-19-parsing-iii-pcfgs-and-inside-outside-algorithm/
PCFG和inside-outside learning（跟inside算法相近但不是一个东西）
https://www.borealisai.com/en/blog/tutorial-15-parsing-i-context-free-grammars-and-cyk-algorithm/
CFG和CYK parsing算法
https://www.borealisai.com/en/blog/tutorial-18-parsing-ii-wcfgs-inside-algorithm-and-weighted-parsing/
WCFG和CYK改良的inside算法
https://www.borealisai.com/en/blog/tutorial-19-parsing-iii-pcfgs-and-inside-outside-algorithm/
PCFG和inside-outside learning（跟inside算法相近但不是一个东西）
Borealisai
Tutorial #18: Parsing II: WCFGs, the inside algorithm, and weighted parsing
We will introduce weighted context-free grammars or WCFGs. These assign a non-negative weight to each rule in the grammar. From here, we can assign a weight to any parse tree by multiplying the weights of its component rules together. We present two variations…
Graph Neural Networks
Foundations, Frontiers, and Applications
Lingfei Wu, JD.COM
Peng Cui, Tsinghua University
Jian Pei, Simon Fraser University
Liang Zhao, Emory University
https://graph-neural-networks.github.io/index.html
Foundations, Frontiers, and Applications
Lingfei Wu, JD.COM
Peng Cui, Tsinghua University
Jian Pei, Simon Fraser University
Liang Zhao, Emory University
https://graph-neural-networks.github.io/index.html
Forwarded from Data Science Archive (小熊猫)
AlibiExplain 应该是这几年看到的在机器学习模型可解释性上做的最系统的工具，堪称知识库型文档，毕竟不能只了解一点 SHAP。https://docs.seldon.io/projects/alibi/en/latest/index.html
https://aclanthology.org/2021.acl-long.379/
https://arxiv.org/abs/2203.00281
给大佬们跪了……比大多数老板都老的CYK算法居然被降到O(n)了
https://arxiv.org/abs/2203.00281
给大佬们跪了……比大多数老板都老的CYK算法居然被降到O(n)了
ACL Anthology
R2D2: Recursive Transformer based on Differentiable Tree for Interpretable Hierarchical Language Modeling
Xiang Hu, Haitao Mi, Zujie Wen, Yafang Wang, Yi Su, Jing Zheng, Gerard de Melo. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long…
https://aaai2022kgreasoning.github.io/
AAAI 2022 Tutorial: Reasoning on Knowledge Graphs: Symbolic or Neural?
一个不错的综述，尽管气氛上这里的“推理”还是计算意义上的，跟认知或心智方面的进路远非字面上那么接近。
AAAI 2022 Tutorial: Reasoning on Knowledge Graphs: Symbolic or Neural?
一个不错的综述，尽管气氛上这里的“推理”还是计算意义上的，跟认知或心智方面的进路远非字面上那么接近。