Combining Predictions under Uncertainty: The Case of Random Decision Trees
Technische Universität Darmstadt(达姆施塔特工业大学);KU Leuven(鲁汶大学)
Florian Busch, Moritz Kulessa, Eneldo Loza Mencía, Hendrik Blockeel
备注:
Preprint version. Appeared in: Discovery Science. 24th International Conference, DS 2021, Halifax, NS, Canada, October 11 to 13, 2021, Proceedings. See https://ds2021.cs.dal.ca/ for further information
发表信息:
In International Conference on Discovery Science (pp. 78-93). Springer, Cham (2021)
A common approach to aggregate classification estimates in an ensemble of decision trees is to either use voting or to average the probabilities for each class. The latter takes uncertainty into account, but not the reliability of the uncertainty estimates (so to say, the "uncertainty about the uncertainty"). More generally, much remains unknown about how to best combine probabilistic estimates from multiple sources. In this paper, we investigate a number of alternative prediction methods. Our methods are inspired by the theories of probability, belief functions and reliable classification, as well as a principle that we call evidence accumulation. Our experiments on a variety of data sets are based on random decision trees which guarantees a high diversity in the predictions to be combined. Somewhat unexpectedly, we found that taking the average over the probabilities is actually hard to beat. However, evidence accumulation showed consistently better results on all but very small leafs.
Time-based Dynamic Controllability of Disjunctive Temporal Networks with Uncertainty: A Tree Search Approach with Graph Neural Network Guidance
NASA Ames Research Center(美国国家航空航天局艾姆斯研究中心);Safran Electronics & Defense(赛峰电子与防务公司);LAMSADE, Paris-Dauphine(巴黎第九大学LAMSADE实验室);Universities Space Research Association(大学空间研究协会);CREC Saint-Cyr Coetquidan(圣西尔-科埃基当军事院校研究中心)
Uncertainty-Aware Extrapolation in Bayesian Oblique Trees
Jožef Stefan Institute(乔泽夫·斯蒂芬研究所);Jožef Stefan International Postgraduate School(乔泽夫·斯蒂芬国际研究生学校);Faculty of Information Studies in Novo mesto(诺沃梅托信息研究学院)
Adaptive Uncertainty-Aware Tree Search for Robust Reasoning
Institute of Software Chinese Academy of Sciences, Beijing, China(中国科学院软件研究所);University of the Chinese Academy of Sciences(中国科学院大学);Multimodal Experiences Lab, Dolby Laboratories Inc(多模态体验实验室,Dolby Laboratories Inc);Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University(人工智能与机器人研究所,西安交通大学);Electronic Information Science and Technology, Central South University(电子信息科学与技术,中南大学)
Beyond Point Prediction: Artificial Representative Trees with Uncertainty
University of Luebeck(吕贝克大学);University Medical Center Göttingen(哥廷根大学医学中心);Lower Saxony Center for Artificial Intelligence and Causal Methods in Medicine(下萨克森州医学人工智能与因果方法中心);Jönköping University(延雪平大学)
Solving Disjunctive Temporal Networks with Uncertainty under Restricted Time-Based Controllability using Tree Search and Graph Neural Networks
NASA Ames Research Center(美国国家航空航天局艾姆斯研究中心);LAMSADE(巴黎多芬大学计算机科学与决策方法实验室);Paris-Dauphine University(巴黎多芬大学);CREC Saint-Cyr Coetquidan(圣西尔科埃基当作战研究中心);Safran(赛峰集团)
Guided Uncertainty-Aware Policy Optimization: Combining Learning and Model-Based Strategies for Sample-Efficient Policy Learning
Stanford University(斯坦福大学);University of California, Berkeley(加州大学伯克利分校);Nvidia(英伟达);University of Toronto(多伦多大学);University of Sydney(悉尼大学);University of Washington(华盛顿大学)
Structural Information Learning Machinery: Learning from Observing, Associating, Optimizing, Decoding, and Abstracting
State Key Laboratory of Software Development Environment(软件开发环境国家重点实验室);School of Computer Science, Beihang University(北京航空航天大学计算机学院)
In the present paper, we propose the model of { \it structural information learning machines} (SiLeM for short), leading to a mathematical definition of learning by merging the theories of computation and information. Our model shows that the essence of learning is { \it to gain information}, that to gain information is { \it to eliminate uncertainty} embedded in a data space, and that to eliminate uncertainty of a data space can be reduced to an optimization problem, that is, an { \it information optimization problem}, which can be realized by a general { \it encoding tree method}. The princi…
Embracing Uncertainty Flexibility: Harnessing a Supervised Tree Kernel to Empower Ensemble Modelling for 2D Echocardiography-Based Prediction of Right Ventricular Volume
Nottingham Trent University(诺丁汉特伦特大学);Columbia University Irving Medical Center(哥伦比亚大学欧文医学中心);University of Leicester(莱斯特大学);Glenfield Hospital(格林菲尔德医院);NIHR Leicester Biomedical Research Centre(NIHR莱斯特生物医学研究中心)