开课单位--华盛顿大学

31
From linearly-solvable optimal control to trajectory optimization, and (hopefully) back[从线性可解最优控制到轨迹优化（希望）返回]
  Emanuel Todorov(华盛顿大学) We have identified a general class of stochastic optimal control problems which are inherently linear, in the sense that the exponentiated optimal val...
热度：89

32
Gaussian processes for Bayesian Filtering[贝叶斯过滤高斯过程]
  Dieter Fox(华盛顿大学)
热度：39

33
Heuristic Search for Generalized Stochastic Shortest Path MDPs[广义随机最短路径问题的启发式搜索]
  Andrey Kolobov(华盛顿大学) Research in efficient methods for solving infinite-horizon MDPs has so far concentrated primarily on discounted MDPs and the more general stochastic s...
热度：77

34
Efficient Weight Learning for Markov Logic Networks[马尔可夫逻辑网络的高效重学习]
  Daniel Lowd(华盛顿大学)
热度：34

35
Combining Logic and Probability: Languages, Algorithms and Applications[结合逻辑和概率：语言，算法和应用]
  Kristian Kersting, Pedro Domingos(华盛顿大学) AI problems are characterized by high degrees of complexity and uncertainty. Complexity is well handled by first-order logic, and uncertainty by proba...
热度：58

36
Function Approximation for Imitation Learning in Humanoid Robots[仿人机器人仿学习功能]
  Rajesh P. N. Rao(华盛顿大学)
热度：33

37
Using Cloud Shadows to Infer Scene Structure and Camera Calibration[使用云阴影推断场景结构和相机校准]
  Nathan Jacobs(华盛顿大学) We explore the use of clouds as a form of structured lighting to capture the 3D structure of outdoor scenes observed over time from a static camera. W...
热度：33

38
Testing and estimation in a sparse normal means model, with connections to shape restricted inference[一个稀疏的正规均值模型的测试和估计，与形状约束推理的连接]
  Jon Wellner(华盛顿大学) Donoho and Jin (2004), following work of Ingster (1999), studied the problem of testing for any signal in a sparse normal means model and showed that ...
热度：37

39
Large-Scale Graph-based Transductive Inference[基于直推式大型图]
  Jeff A. Bilmes(华盛顿大学) We consider the issue of scalability of graph-based semi-supervised learning (SSL) algorithms. In this context, we propose a fast graph node ordering ...
热度：25

40
Directed Graph Embedding: an Algorithm based on Continuous Limits of Laplacian-type Operators[有向图的嵌入：基于拉普拉斯型算子的连续极限算法]
  Dominique Perrault-Joncas(华盛顿大学) This paper considers the problem of embedding directed graphs in Euclidean space while retaining directional information. We model the observed graph ...
热度：46