Proximal and projected gradient descent.Gradient and subgradient descent: Lipschitz functions, Smooth functions, Smooth and Strongly Convex functions.First order optimality conditions for unconstrained and constrained convex optimization problems.Convex sets and Convex functions, including basic definitions of convexity, smoothness and strong convexity.Topics include growth of functions, divide-and-conquer algorithms, dynamic programming, greedy algorithms, basic graph algorithms, network flow, minimum-cost matching, linear program-ming, randomized algorithms, data structures (hashing, amortized analysis, splay trees, union-find, and Fibonacci heaps), online algorithms for paging, P, NP, NP-completeness, and approximation algorithms. This course presents techniques for the design and analysis of efficient approximation algorithms. Even if a given optimization problem is NP-hard, it may be possible to compute near-optimal solutions efficiently. This course presents techniques for establishing evidence of such computational intractability, especially NP-hardness. ![]() Unfortunately, many optimization problems that arise in practice are unlikely to be polynomial-time solvable. This course presents techniques for the design and analysis of polynomial-time algorithms. In this setting, it is crucial to employ asymptotically efficient algorithms. Modern computational applications often involve massive data sets. Professors Peter Stone and Scott Niekum are active reinforcement learning researchers and bring their expertise and excitement for RL to the class. The material covered in this class will provide an understanding of the core fundamentals of reinforcement learning, preparing students to apply it to problems of their choosing, as well as allowing them to understand modern RL research. Reinforcement learning is an essential part of fields ranging from modern robotics to game-playing (e.g. It covers the essentials of reinforcement learning (RL) theory and how to apply it to real-world sequential decision problems. The course will cover model-free and model-based reinforcement learning methods, especially those based on temporal difference learning and policy gradient algorithms. ![]() Reinforcement learning problems involve learning what to do-how to map situations to actions-so as to maximize a numerical reward signal. ![]() Introduces the theory and practice of modern reinforcement learning. This course introduces the theory and practice of modern reinforcement learning.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |