惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Application and Cybersecurity Blog
Application and Cybersecurity Blog
S
Securelist
K
Kaspersky official blog
Scott Helme
Scott Helme
C
CXSECURITY Database RSS Feed - CXSecurity.com
GbyAI
GbyAI
钛媒体：引领未来商业与生活新知
钛媒体：引领未来商业与生活新知
C
Cisco Blogs
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
博
博客园 - Franky
Security Latest
Security Latest
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Y
Y Combinator Blog
T
Threat Research - Cisco Blogs
L
LINUX DO - 热门话题
C
Cyber Attacks, Cyber Crime and Cyber Security
Project Zero
Project Zero
Cisco Talos Blog
Cisco Talos Blog
月光博客
月光博客
I
Intezer
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
人人都是产品经理
人人都是产品经理
L
Lohrmann on Cybersecurity
Recorded Future
Recorded Future
Latest news
Latest news
V2EX - 技术
V2EX - 技术
T
The Exploit Database - CXSecurity.com
H
Heimdal Security Blog
F
Fortinet All Blogs
Cloudbric
Cloudbric
IT之家
IT之家
博
博客园 - 叶小钗
Microsoft Security Blog
Microsoft Security Blog
P
Proofpoint News Feed
博
博客园 - 司徒正美
Apple Machine Learning Research
Apple Machine Learning Research
PCI Perspectives
PCI Perspectives
AWS News Blog
AWS News Blog
H
Help Net Security
S
Security @ Cisco Blogs
酷 壳 – CoolShell
酷 壳 – CoolShell
Recent Announcements
Recent Announcements
Hacker News - Newest:
Hacker News - Newest: "LLM"
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
F
Full Disclosure
S
Schneier on Security
S
Security Affairs
T
Tenable Blog

stat.ML updates on arXiv.org

Adaptive multi-fidelity optimization with fast learning rates Enhancing AI and Dynamical Subseasonal Forecasts with Probabilistic Bias Correction Sample Complexity Bounds for Stochastic Shortest Path with a Generative Model The Harder Path: Last Iterate Convergence for Uncoupled Learning in Zero-Sum Games with Bandit Feedback Stylistic-STORM (ST-STORM) : Perceiving the Semantic Nature of Appearance Collective Kernel EFT for Pre-activation ResNets PRIM-cipal components analysis One-Shot Generative Flows: Existence and Obstructions Structural interpretability in SVMs with truncated orthogonal polynomial kernels Amortized Optimal Transport from Sliced Potentials MinShap: A Modified Shapley Value Approach for Feature Selection Unsupervised feature selection using Bayesian Tucker decomposition Multi-User mmWave Beam and Rate Adaptation via Combinatorial Satisficing Bandits Best of both worlds: Stochastic & adversarial best-arm identification Scalable Model-Based Clustering with Sequential Monte Carlo Expert-Guided Class-Conditional Goodness-of-Fit Scores for Interpretable Classification with Informative Missingness: An Application to Seismic Monitoring Lightweight Geometric Adaptation for Training Physics-Informed Neural Networks Gating Enables Curvature: A Geometric Expressivity Gap in Attention Zeroth-Order Optimization at the Edge of Stability Differentially Private Conformal Prediction CLion: Efficient Cautious Lion Optimizer with Enhanced Generalization Generative Augmented Inference Improving Machine Learning Performance with Synthetic Augmentation PAC-MCTS: Bias-Aware Pruning for Robust LLM-Guided Search and Planning Path-Sampled Integrated Gradients Heat and Matérn Kernels on Matchings Doubly Outlier-Robust Online Infinite Hidden Markov Model Momentum Further Constrains Sharpness at the Edge of Stochastic Stability Multistage Conditional Compositional Optimization BOAT: Navigating the Sea of In Silico Predictors for Antibody Design via Multi-Objective Bayesian Optimization Sandpile Economics: Theory, Identification, and Evidence Online learning with noisy side observations Spectral Thompson sampling Covariance-adapting algorithm for semi-bandits with application to sparse rewards Ordinary Least Squares is a Special Case of Transformer Metric-Aware Principal Component Analysis (MAPCA):A Unified Framework for Scale-Invariant Representation Learning Robust Low-Rank Tensor Completion based on M-product with Weighted Correlated Total Variation and Sparse Regularization Joint Representation Learning and Clustering via Gradient-Based Manifold Optimization Universality of Gaussian-Mixture Reverse Kernels in Conditional Diffusion Interpretable and Explainable Surrogate Modeling for Simulations: A State-of-the-Art Survey and Perspectives on Explainable AI for Decision-Making Estimating Continuous Treatment Effects with Two-Stage Kernel Ridge Regression A short proof of near-linear convergence of adaptive gradient descent under fourth-order growth and convexity Some Theoretical Limitations of t-SNE Bias-Corrected Adaptive Conformal Inference for Multi-Horizon Time Series Forecasting Identifiability of Potentially Degenerate Gaussian Mixture Models With Piecewise Affine Mixing Rare Event Analysis via Stochastic Optimal Control Adaptive Learning via Off-Model Training and Importance Sampling for Fully Non-Markovian Optimal Stochastic Control. Complete version Generalization Guarantees on Data-Driven Tuning of Gradient Descent with Langevin Updates Minimizing classical resources in variational measurement-based quantum computation for generative modeling Deep Learning for Sequential Decision Making under Uncertainty: Foundations, Frameworks, and Frontiers ADD for Multi-Bit Image Watermarking Beyond Fixed False Discovery Rates: Post-Hoc Conformal Selection with E-Variables Regional Explanations: Bridging Local and Global Variable Importance ShapShift: Explaining Model Prediction Shifts with Subgroup Conditional Shapley Values Cost-optimal Sequential Testing via Doubly Robust Q-learning Query Lower Bounds for Diffusion Sampling Tail-Aware Information-Theoretic Generalization for RLHF and SGLD Beyond Augmented-Action Surrogates for Multi-Expert Learning-to-Defer Hierarchical Kernel Transformer: Multi-Scale Attention with an Information-Theoretic Approximation Analysis Policy-Aware Design of Large-Scale Factorial Experiments Towards Verified and Targeted Explanations through Formal Methods Portfolio Optimization Proxies under Label Scarcity and Regime Shifts via Bayesian and Deterministic Students under Semi-Supervised Sandwich Training Spectral methods: crucial for machine learning, natural for quantum computers? The Devil Is in Gradient Entanglement: Energy-Aware Gradient Coordinator for Robust Generalized Category Discovery A Tutorial Review of Bayesian Optimization with Gaussian Processes to Accelerate Stationary Point Searches Certified and accurate computation of function space norms of deep neural networks Mini-Batch Covariance, Diffusion Limits, and Oracle Complexity in Stochastic Gradient Descent: A Sampling-Design Perspective Conformal Policy Control Diagnostics for Individual-Level Prediction Instability in Machine Learning for Healthcare Neural Networks With Dense Weights Are Not Universal Approximators Continuous-time reinforcement learning: ellipticity enables model-free value function approximation Scalable spatial point process models for forensic footwear analysis A Review of Diffusion-based Simulation-Based Inference: Foundations and Applications in Non-Ideal Data Scenarios Active Learning with Selective Time-Step Acquisition for PDEs Joint Score-Threshold Optimization for Interpretable Risk Assessment Revisiting Entropy Regularization: Adaptive Coefficient Unlocks Its Potential for LLM Reinforcement Learning Discrete Guidance Matching: Exact Guidance for Discrete Flow Matching PnP-CM: Consistency Models as Plug-and-Play Priors for Inverse Problems Online Distributionally Robust LLM Alignment via Regression to Relative Reward Heavy-Tailed Class-Conditional Priors for Long-Tailed Generative Modeling Random Walk Learning and the Pac-Man Attack Sequential Regression Learning with Randomized Algorithms Diagnosing and Improving Diffusion Models by Estimating the Optimal Loss Value Random Matrix Theory for Deep Learning: Beyond Eigenvalues of Linear Models Scalable Spatiotemporal Inference with Biased Scan Attention Transformer Neural Processes Towards AI-assisted Neutrino Flavor Theory Design Towards Reasonable Concept Bottleneck Models Practical estimation of the optimal classification error with soft labels and calibration Flow-based Generative Modeling of Potential Outcomes and Counterfactuals The Gaussian Latent Machine: Efficient Prior and Posterior Sampling for Inverse Problems Two-Dimensional Deep ReLU CNN Approximation for Korobov Functions: A Constructive Approach FSPO: Few-Shot Optimization of Synthetic Preferences Personalizes to Real Users Identifying Information from Observations with Uncertainty and Novelty A ghost mechanism: An analytical model of abrupt learning in recurrent networks A Multiparty Homomorphic Encryption Approach to Confidential Federated Kaplan Meier Survival Analysis Large Language Models for Market Research: A Data-augmentation Approach Transformer Neural Processes - Kernel Regression FIT-GNN: Faster Inference Time for GNNs that 'FIT' in Memory Using Coarsening Estimating Joint Interventional Distributions from Marginal Interventional Data Nonparametric Sparse Online Learning of the Koopman Operator
Exploration-Exploitation Trade-off in Reinforcement Learning on Online Markov Decision Processes with Global Concave Rewards
Wang Chi Cheung · 2019-05-16 · via stat.ML updates on arXiv.org

We consider an agent who is involved in a Markov decision process and receives a vector of outcomes every round. Her objective is to maximize a global concave reward function on the average vectorial outcome. The problem models applications such as multi-objective optimization, maximum entropy exploration, and constrained optimization in Markovian environments. In our general setting where a stationary policy could have multiple recurrent classes, the agent faces a subtle yet consequential trade-off in alternating among different actions for balancing the vectorial outcomes. In particular, stationary policies are in general sub-optimal. We propose a no-regret algorithm based on online convex optimization (OCO) tools (Agrawal and Devanur 2014) and UCRL2 (Jaksch et al. 2010). Importantly, we introduce a novel gradient threshold procedure, which carefully controls the switches among actions to handle the subtle trade-off. By delaying the gradient updates, our procedure produces a non-stationary policy that diversifies the outcomes for optimizing the objective. The procedure is compatible with a variety of OCO tools.

此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。