V
主页
【Proof-Trivial】深度神经网络中的【泛化理论】与【归纳偏差】 (Generalization and Inductive Bias)
发布人
https://www.youtube.com/@communicationsandsignalpro6570 Cengiz Pehlevan Assistant Professor of Applied Mathematics Harvard University ABSTRACT: I will present a theory that describes generalization and inductive bias in neural networks using kernel methods and statistical mechanics. This theory accurately predicts the generalization performance of neural networks and generic kernels on real data, and elucidates an inductive bias to explain data with “simple functions”, which are identified by solving a related kernel eigenfunction problem on the data distribution. This notion of simplicity allows us to characterize whether a network is compatible with a learning task, facilitating good generalization performance from a small number of training examples. I will present applications of this theory to artificial and biological neural systems, and real datasets. I will discuss extensions to out-of-distribution generalization and data-dependent kernel descriptions of neural networks. References: https://proceedings.mlr.press/v119/bo... https://www.nature.com/articles/s4146... https://arxiv.org/abs/2106.02261 https://arxiv.org/abs/2111.00034 BIO: Cengiz (pronounced “Jen-ghiz”) Pehlevan is an assistant professor of applied mathematics at Harvard SEAS. His research interests are in theoretical neuroscience and theory of neural computation. Cengiz comes to Harvard SEAS from the Flatiron Institute’s Center for Computational Biology (CCB), where he was a a research scientist in the neuroscience group. Before CCB, Cengiz was a postdoctoral associate at Janelia Research Campus, and before that a Swartz Fellow at Harvard. Cengiz received a doctorate in physics from Brown University and undergraduate degrees in physics and electrical engineering from Bogazici University, Turkey.
打开封面
下载高清视频
观看高清视频
视频下载器
【Proof-Trivial】控制理论中的在线优化(Competitive Control via Online Optimization) @加州理工学院
【Proof-Trivial】随机矩阵理论
【Proof-Trivial】黎曼几何【北京大学/北师大 葛剑】
【Proof-Trivial】 Geometric Optimization:黎曼流形上的加速梯度法—— Suvrit Sra (MIT LIDS)
B站强推!2024公认最通俗易懂的【Transformer】教程,125集付费课程(附资料)神经网络_注意力机制_深度学习_BERT_大模型
【Proof-Trivial】流形入门 (Manifolds)
【Proof-Trivial】当【深度学习】遇见【稀疏正则化】 —— 信号处理视角
【Proof-Trivial】量子力学与高等量子力学【北京大学 郭弘】更新中......
【Proof-Trivial】最优传输前沿讲座 (Optimal Transport) (更新中...)
哈佛大学惊人发现:所有的疾病,最怕这一个字,再不看不懂就晚了
【Proof-Trivial】机器学习与数据科学中的【理论】与【算法】【北京大学 文再文】
【Proof-Trivial】分布鲁棒优化 (Distributionally Robust Optimization, DRO)
【Proof-Trivial】K-理论 (K-theory)
【Proof-Trivial】【入门】基于物理信息的神经网络-Physics-Informed Neural Networks (PINNs)
【Proof-Trivial】【2023年】最优化:建模、算法与理论 【北京大学 文再文】
【Proof-Trivial】高等数理统计(基于测度论) 清华大学
【Proof-Trivial】通信与感知 —— 从【压缩采样】到【基于模型的深度学习】
【Proof-Trivial】高维概率及其在数据科学中的应用 (HDP) - Roman Vershynin
【Proof-Trivial】【讲座】在线学习与在线凸优化 (Online Learning&Convex Optimization)
【Proof-Trivial】强化学习中的自适应离散化 (Adaptive Discretization For Reinforcement Learning)
【Proof-Trivial】量子信号处理 (Quantum Signal Processing)
【Proof-Trivial】贝叶斯优化与贝叶斯学习 (Bayesian Optimization) 持续更新
【Proof-Trivial】深度强化学习基础 (DRL)——复旦大学 (赵卫东)
【Proof-Trivial】从高斯过程(GP)到神经正切核(NTK) —— 无限宽神经网络指北
【Proof-Trivial】【IROS‘22】机器人学习、优化与控制中的几何方法
【Proof-Trivial】泛函分析保姆级入门教程 (Functional Analysis)
【Proof-Trivial】哈密顿系统与辛几何
【Proof-Trivial】信息几何【入门课程】 (Melvin Leok from UCSD)
【Proof-Trivial】变分学 【北京大学 张恭庆院士】【希尔伯特第23个问题】
【Proof-Trivial】李理论 (Lie theory) —— 李群、李代数、李括号
【全195集】禁止自学走弯路!回归算法、聚类算法、决策树、随机森林、神经网络、贝叶斯算法、支持向量机等十大机器学习算法一口气学完!
【Proof-Trivial】马尔科夫链及其算法应用 (MCAA, EPFL)
超全超简单!入门到进阶一口气学完CNN、RNN、GAN、GNN、DQN、Transformer、LSTM等七大深度学习神经网络算法!真的比啃书好太多!
【Proof-Trivial】深度学习理论讲座 (Deep Learning Theory) 持续更新......
【Proof-Trivial】离线批次强化学习 —— 从历史中学到更多 (Offline Batch RL)
【Proof-Trivial】数学视角下的Transformer【MIT数学系-Philippe Rigollett】
【Proof-Trivial】现代机器学习中的【统计学习理论】
【Proof-Trivial】将Transformers视为支持向量机 (Transformers as Support Vector Machines)
【Proof-Trivial】统计热力学 北京大学
【Proof-Trivial】实验设计前沿讲座 (Experimental Design)