no code implementations • 31 Aug 2023 • Bobak Pezeshki, Radu Marinescu, Alexander Ihler, Rina Dechter
Scientific computing has experienced a surge empowered by advancements in technologies such as neural networks.
no code implementations • 7 Oct 2022 • Noble Kennamer, Steven Walton, Alexander Ihler
Bayesian optimal experimental design is a sub-field of statistics focused on developing methods to make efficient use of experimental resources.
1 code implementation • 16 Sep 2022 • Litian Liang, Yaosheng Xu, Stephen Mcaleer, Dailin Hu, Alexander Ihler, Pieter Abbeel, Roy Fox
On a set of 26 benchmark Atari environments, MeanQ outperforms all tested baselines, including the best available baseline, SUNRISE, at 100K interaction steps in 16/26 environments, and by 68% on average.
no code implementations • AAAI Workshop CLeaR 2022 • Sakshi Agarwal, Kalev Kask, Alexander Ihler, Rina Dechter
A major limiting factor in graphical model inference is the complexity of computing the partition function.
no code implementations • 28 Oct 2021 • Litian Liang, Yaosheng Xu, Stephen Mcaleer, Dailin Hu, Alexander Ihler, Pieter Abbeel, Roy Fox
Under the belief that $\beta$ is closely related to the (state dependent) model uncertainty, Entropy Regularized Q-Learning (EQL) further introduces a principled scheduling of $\beta$ by maintaining a collection of the model parameters that characterizes model uncertainty.
1 code implementation • 12 Oct 2020 • Noble Kennamer, Emille E. O. Ishida, Santiago Gonzalez-Gaitan, Rafael S. de Souza, Alexander Ihler, Kara Ponder, Ricardo Vilalta, Anais Moller, David O. Jones, Mi Dai, Alberto Krone-Martins, Bruno Quint, Sreevarsha Sreejith, Alex I. Malz, Lluis Galbany
The Recommendation System for Spectroscopic follow-up (RESSPECT) project aims to enable the construction of optimized training samples for the Rubin Observatory Legacy Survey of Space and Time (LSST), taking into account a realistic description of the astronomical data environment.
no code implementations • ECCV 2018 • Shaofei Wang, Alexander Ihler, Konrad Kording, Julian Yarkony
We present a novel approach to solve dynamic programs (DP), which are frequent in computer vision, on tree-structured graphs with exponential node state space.
no code implementations • ICML 2018 • Noble Kennamer, David Kirkby, Alexander Ihler, Francisco Javier Sanchez-Lopez
Our particular motivation is star galaxy classification for ground based optical surveys.
no code implementations • NeurIPS 2016 • Wei Ping, Qiang Liu, Alexander Ihler
In this work, we propose an infinite restricted Boltzmann machine~(RBM), whose maximum likelihood estimation~(MLE) corresponds to a constrained convex optimization.
no code implementations • 18 Sep 2017 • Shaofei Wang, Chong Zhang, Miguel A. Gonzalez-Ballester, Alexander Ihler, Julian Yarkony
We give a novel integer program formulation of the multi-person pose estimation problem, in which variables correspond to assignments of parts in the image to poses in a two-tier, hierarchical way.
no code implementations • 2 Mar 2017 • Wei Ping, Alexander Ihler
We demonstrate that, in both maximum likelihood and max-margin learning, training conditional RBMs with BP as the inference routine can provide significantly better results than current state-of-the-art CD methods on structured prediction problems.
no code implementations • NeurIPS 2015 • Wei Ping, Qiang Liu, Alexander Ihler
Marginal MAP inference involves making MAP predictions in systems defined with latent variables or missing information.
no code implementations • NeurIPS 2014 • Qiang Liu, Alexander Ihler
Distributed learning of probabilistic models from multiple data repositories with minimum communication is increasingly important.
no code implementations • 4 Sep 2014 • Wei Ping, Qiang Liu, Alexander Ihler
In this work, we propose the marginal structured SVM (MSSVM) for structured prediction with hidden variables.
no code implementations • 26 Feb 2013 • Qiang Liu, Alexander Ihler
The marginal maximum a posteriori probability (MAP) estimation problem, which calculates the mode of the marginal posterior distribution of a subset of variables with the remaining variables marginalized, is an important inference problem in many models, such as those with hidden variables or uncertain parameters.