- 博客(20)
- 收藏
- 关注
原创 RL--Policy-based methods--PPO
policy gradient 的 off-policy 版本,有 2 个 policy,behavior policy 用来采样,target policy 是训练最终要得到的结果。PPO全称Proximal Policy Optimizationproximal 为近端的,表示要让 behavior policy 和 target policy 差别不太大。虽然采样和训练的2 个 policy,但是需要他们差别较小。
2023-10-13 00:13:30 117
原创 RL--Policy-based methods--Policy Gradient
通过 gradient ascent 让 policy 下 trajectory 的期望 return 变得尽可能的大。
2023-10-12 15:21:41 91
原创 大模型--高效微调--Prefix Tuning
给每个任务初始化一些virtual token 的 embedding 作为 prefix 添加到序列中,训练时只优化 virtual token embedding只在输入时把 task 的 token embedding 加入输入序列的 embedding。
2023-09-23 18:09:13 294
原创 大模型-高效微调-LoRA
freeze the pre-trained model weights, inject trainable low-rank matrices into each layer, 从而只用训练很少的参数,就可以达到全参数微调一样的效果。
2023-09-23 16:32:15 86
原创 GNN-CS224W: 19 GNNs for Science
Pre-Training GNNspre-train要考虑的内容pre-train之前可以先设计好希望训练完的embedding满足什么要求,例如同类型的要接近,不同类型的要距离远,有相同属性的要接近,推理的要求等等pre-train taskself-supervised 和superviseembedding的level例如node level,graph level,link levelTasksAttribute Masking [Node-level] [self
2022-03-20 10:48:45 154
原创 GNN-CS224W: 17 Scaling Up GNNs
Graphs in model applicationapplicationsRecommender systems:Amazon, YouTube 等要预测用户对item(商品、视频)是否感兴趣 (link prediction)还要预测用户、item的类型(node classification)数据量级很大Social networksFacebook, Twitter, Instagram等要做 Friend recommendation (link level task)Us
2022-03-20 10:48:20 502
原创 GNN-CS224W: 16 Position-aware and Identity-aware GNNs and Robustness of GNN
The GNNs so far are not perfectwhat should a perfect GNN do ?A perfect GNN should build an injective function between neighborhood structure (regardless of hops) and node embeddings.根据以前讲的知识,a perfect GNN 应该满足2点:If two nodes have the same neighborhood
2022-03-20 10:47:56 911
原创 GNN-CS224W:13 Community Structure in Networks
connection between edge strength and edge structureedge strength在social network中可以用两个人之间联系的频率来得到,联系的频率越高,则strength越大,否则越小edge structure表示这个edge是处于一个cluster内部,还是处于不同cluster之间edge overlap用edge overlap来表示edge structureOij=∣(N(i)∩N(j))−{i,j}∣∣(N(i)∪N(j)
2022-03-20 10:43:43 148
原创 GNN-CS224W:12 Frequent Subgraph Mining with GNNs
Subgraphs and motifs本节讲的是如何定义subgraph和motif,以及如何用subgraph和motif 来代表graph的信息。NotionsSubgraph DefinitionNode-induced subgraph ( induced subgraph ): Take subset of the nodes and all edges induced by the nodes,all edges指的是edge的两个节点都在node subset里的edgeEdge-
2022-03-19 20:09:26 363
原创 GNN-CS224W: 11 Reasoning over Knowledge Graphs
QueryOne-hop Queries转化为KG completion taskKG completion 实现对候选tail做分类,即(head,relation,tail)(head, relation, tail)(head,relation,tail)是否存在问题,怎么确定候选tail?难道所有的都要试一遍?Path Queriesn hop query 可以表示为q=(va,(r1,…,rn))q=(v_a, (r_1, \dots, r_n))q=(va,(r1,…,rn
2022-03-19 20:08:56 157
原创 GNN-CS224W:10 Knowledge Graph Embeddings
Heterogeneous Graph有不止一种edge、node的graph就是Heterogeneous GraphHeterogeneous Graph可以用4元组来表示:G=(V,E,R,T)G=(V,E,R,T)G=(V,E,R,T)其中vi∈Vv_i \in Vvi∈V表示node,node 的type可以通过T(vi)T(v_i)T(vi)得到edge用3元组表示(vi,r,vj)∈E(v_i, r, v_j) \in E(vi,r,vj)∈E,关系r∈Rr \in Rr∈R
2022-03-19 20:08:33 229
原创 GNN-CS224W: 9 Theory of Graph Neural Networks
What is the expressive power(ability to distinguish different graph structures) of these GNN models?graph包括2部分:1. node feature;2. graph structure接下来先讨论graph structure,假设node feature都一样(是否会讨论node featue 不知道,接着看)接下来讨论的问题是:How well can a GNN distinguish d
2022-03-19 20:07:58 257
原创 GNN-CS224W: 8 Training GNN
Graph AugmentationRaw input graph ≠ computational graph之前提到的computational graph是完全根据node之间关系确定的,但是不一定非得这样做,可以通过其他方法获得computational graph为什么不用原始的computational graph?因为原始的computational graph存在各种问题:The input graph lacks featuresGraph structure problem
2022-03-19 20:07:37 439
原创 GNN-CS224W: 6-7 Graph Neural Networks
GNN概述GNN定义了一种可以批量encode graph 类型数据的方法的深度方法可以利用local graph structure,也可以利用node featurecomputational graph如下图所示:node 的local structure 定义了它的computational graph每个节点都计算相同的depthnode的输入就是feature,link体现在node 的computational graph上同一个node在不同的层会有不同的embedd
2022-03-19 20:07:13 147
原创 GNN-CS224W: 5 Label Propagation for Node Classification
Task全文都在解决这一个Task: Given a network with labels on some nodes, how do we assign labels to all other nodes in the network?本文涉及的是Probabilistic framework,不涉及深度学习。用到的信息Classification label of a node vvv in network may depend onFeatures of node vvvLabels
2022-03-19 20:06:51 177
原创 GNN-CS224W: 4 Link Analysis: PageRank
treat the web as a graphpage rank的目的是给作为node的网页排序We will cover the following Link Analysis approaches to compute the importance of nodes in a graph:PageRankPersonalized PageRank (PPR)Random Walk with Restarts3种方法是类似的,是逐渐改进的。目的是计算node的importanceIde
2022-03-19 20:06:07 229
原创 GNN-CS224W: 3 Node Embeddings
graph representation learningAutomaticly learning efficient task-independent feature for machine learning with graphs,从而避免了特征工程why embedding?Similarity of embeddings between nodes indicates their similarity in the network.Encode network informationTa
2022-03-19 20:04:18 404
原创 GNN-CS224W: 1-2 Introduction; Traditional Methods for machine learning in Graphs
node-level, link-level, graph-level tasks and features
2022-03-19 20:03:11 974
空空如也
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人