1 code implementation • NeurIPS 2013 • Nan Ye, Adhiraj Somani, David Hsu, Wee Sun Lee
We show that the best policy obtained from a DESPOT is near-optimal, with a regret bound that depends on the representation size of the optimal policy.
1 code implementation • ICLR 2020 • Xiao Ma, Peter Karkus, David Hsu, Wee Sun Lee, Nan Ye
The particle filter maintains a belief using learned discriminative update, which is trained end-to-end for decision making.
1 code implementation • 13 Oct 2018 • Drew Mitchell, Nan Ye, Hans De Sterck
While Nesterov acceleration turns gradient descent into an optimal first-order method for convex problems by adding a momentum term with a specific weight sequence, a direct application of this method and weight sequence to ALS results in erratic convergence behaviour.
1 code implementation • 1 Dec 2020 • Aaron J. Snoswell, Surya P. N. Singh, Nan Ye
This improves the previous heuristic derivation of the MaxEnt IRL model (for stochastic MDPs), allows a unified view of MaxEnt IRL and Relative Entropy IRL, and leads to a model-free learning algorithm for the MaxEnt IRL model.
1 code implementation • 16 Oct 2022 • Jonathan Wilton, Abigail M. Y. Koay, Ryan K. L. Ko, Miao Xu, Nan Ye
Key to our approach is a new interpretation of decision tree algorithms for positive and negative data as \emph{recursive greedy risk minimization algorithms}.
1 code implementation • 13 Sep 2022 • Marcus Hoerger, Hanna Kurniawati, Dirk Kroese, Nan Ye
A Voronoi tree is a Binary Space Partitioning (BSP) that implicitly maintains the partition of a cell as the Voronoi diagram of two points sampled from the cell.
1 code implementation • 21 Feb 2023 • Marcus Hoerger, Hanna Kurniawati, Dirk Kroese, Nan Ye
ADVT uses the estimated diameters of the cells to form an upper-confidence bound on the action value function within the cell, guiding the Monte Carlo Tree Search expansion and further discretization of the action space.
1 code implementation • 20 Nov 2023 • Wei Jiang, Zhuang Xiong, Feng Liu, Nan Ye, Hongfu Sun
Supervised deep learning methods have shown promise in undersampled Magnetic Resonance Imaging (MRI) reconstruction, but their requirement for paired data limits their generalizability to the diverse MRI acquisition parameters.
1 code implementation • 20 Dec 2023 • Jonathan Wilton, Nan Ye
We consider training decision trees using noisily labeled data, focusing on loss functions that can lead to robust learning algorithms.
1 code implementation • 9 Oct 2019 • Tan Nguyen, Nan Ye, Peter L. Bartlett
Theoretically, we first consider whether we can use linear, instead of convex, combinations, and obtain generalization results similar to existing ones for learning from a convex hull.
no code implementations • 30 Mar 2016 • Nguyen Viet Cuong, Nan Ye, Wee Sun Lee
This suggests we should use a Lipschitz utility for AL if robustness is required.
no code implementations • NeurIPS 2013 • Nguyen Viet Cuong, Wee Sun Lee, Nan Ye, Kian Ming A. Chai, Hai Leong Chieu
We introduce a new objective function for pool-based Bayesian active learning with probabilistic hypotheses.
no code implementations • NeurIPS 2009 • Nan Ye, Wee S. Lee, Hai L. Chieu, Dan Wu
Dependencies among neighbouring labels in a sequence is an important source of information for sequence labeling problems.
no code implementations • ICML 2017 • Andrew Wrigley, Wee Sun Lee, Nan Ye
We propose a new approximate inference algorithm for graphical models, tensor belief propagation, based on approximating the messages passed in the junction tree algorithm.
no code implementations • 3 Jun 2021 • Aaron J. Snoswell, Surya P. N. Singh, Nan Ye
Multiple-Intent Inverse Reinforcement Learning (MI-IRL) seeks to find a reward function ensemble to rationalize demonstrations of different but unlabelled intents.
no code implementations • 19 May 2022 • Yawen Zhao, Mingzhe Zhang, Chenhao Zhang, Weitong Chen, Nan Ye, Miao Xu
This is because AdaPU learns a weak classifier and its weight using a weighted positive-negative (PN) dataset with some negative data weights $-$ the dataset is derived from the original PU data, and the data weights are determined by the current weighted classifier combination, but some data weights are negative.
1 code implementation • 14 May 2023 • Marcus Hoerger, Hanna Kurniawati, Dirk Kroese, Nan Ye
At each planning step, our method uses a novel lazy Cross-Entropy method to search the space of policy trees, which provide a simple policy representation.