variational lower bound
Connecting Jensen–Shannon and Kullback–Leibler Divergences: A New Bound for Representation Learning
Mutual Information (MI) is a fundamental measure of statistical dependence widely used in representation learning. While direct optimization of MI via its definition as a Kullback-Leibler divergence (KLD) is often intractable, many recent methods have instead maximized alternative dependence measures, most notably, the Jensen-Shannon divergence (JSD) between joint and product of marginal distributions via discriminative losses. However, the connection between these surrogate objectives and MI remains poorly understood.
Partial Multi-Label Learning with Probabilistic Graphical Disambiguation
In partial multi-label learning (PML), each training example is associated with a set of candidate labels, among which only some labels are valid. As a common strategy to tackle PML problem, disambiguation aims to recover the ground-truth labeling information from such inaccurate annotations. However, existing approaches mainly rely on heuristics or ad-hoc rules to disambiguate candidate labels, which may not be universal enough in complicated real-world scenarios. To provide a principled way for disambiguation, we make a first attempt to explore the probabilistic graphical model for PML problem, where a directed graph is tailored to infer latent ground-truth labeling information from the generative process of partial multi-label data. Under the framework of stochastic gradient variational Bayes, a unified variational lower bound is derived for this graphical model, which is further relaxed probabilistically so that the desired prediction model can be induced with simultaneously identified ground-truth labeling information. Comprehensive experiments on multiple synthetic and real-world data sets show that our approach outperforms the state-of-the-art counterparts.
Variational Autoencoder for Deep Learning of Images, Labels and Captions
Yunchen Pu, Zhe Gan, Ricardo Henao, Xin Yuan, Chunyuan Li, Andrew Stevens, Lawrence Carin
Neural Information Processing Systems http://nips.cc/
Variational Information Maximization for Feature Selection
Shuyang Gao, Greg Ver Steeg, Aram Galstyan
Neural Information Processing Systems http://nips.cc/
Cluster Variational Approximations for Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data
Neural Information Processing Systems http://nips.cc/
Probabilistic Model-Agnostic Meta-Learning
Chelsea Finn, Kelvin Xu, Sergey Levine
In this paper,we propose aprobabilisticmeta-learning algorithm that can sample models for a new task from a modeldistribution.
Supplementary Material for "SE(3) Diffusion Model-based Point Cloud Registration for Robust 6D Object Pose Estimation "
SE(3) diffusion model for point cloud registration can be derived as below. By inserting Eq. 5 into the variational lower bound 4, we can further rewrite the variational lower As demonstrated in our main paper, we utilize the Lie algebra for randomly sampling the desired perturbation transformation to randomize our SE(3) diffusion process. This innovative registration framework exhibits promising registration performance. Learning 6d object pose estimation using 3d object coordinates.
