Search Results for author:
Found 18 papers, 9 papers with code
Improving Transformer World Models for Data-Efficient RL
We present an approach to model-based RL that achieves a new state of the art performance on the challenging Craftax-classic benchmark, an open-world 2D survival game that requires agents to exhibit a wide range of general abilities -- such as strong generalization, deep exploration, and long-term reasoning.
DMC-VB: A Benchmark for Representation Learning for Control with Visual Distractors
In this paper, we present theDeepMind Control Visual Benchmark (DMC-VB), a dataset collected in the DeepMind Control Suite to evaluate the robustness of offline RL agents for solving continuous control tasks from visual input in the presence of visual distractors.
Learning Cognitive Maps from Transformer Representations for Efficient Planning in Partially Observed Environments
Despite their stellar performance on a wide range of tasks, including in-context tasks only revealed during inference, vanilla transformers and variants trained for next-token predictions (a) do not learn an explicit world model of their environment which can be flexibly queried and (b) cannot be used for planning or navigation.
Learning noisy-OR Bayesian Networks with Max-Product Belief Propagation
We evaluate both approaches on several benchmarks where VI is the state-of-the-art and show that our method (a) achieves better test performance than Ji et al. (2020) for learning noisy-OR BNs with hierarchical latent structures on large sparse real datasets; (b) recovers a higher number of ground truth parameters than Buhai et al. (2020) from cluttered synthetic scenes; and (c) solves the 2D blind deconvolution problem from Lazaro-Gredilla et al. (2021) and variant - including binary matrix factorization - while VI catastrophically fails and is up to two orders of magnitude slower.
PGMax: Factor Graphs for Discrete Probabilistic Graphical Models and Loopy Belief Propagation in JAX
PGMax is an open-source Python package for (a) easily specifying discrete Probabilistic Graphical Models (PGMs) as factor graphs; and (b) automatically running efficient and scalable loopy belief propagation (LBP) in JAX.
Graphical Models with Attention for Context-Specific Independence and an Application to Perceptual Grouping
To demonstrate MAM's capabilities to capture CSIs at scale, we apply MAMs to capture an important type of CSI that is present in a symbolic approach to recurrent computations in perceptual grouping.
Perturb-and-max-product: Sampling and learning in discrete energy-based models
Perturb-and-MAP offers an elegant approach to approximately sample from a energy-based model (EBM) by computing the maximum-a-posteriori (MAP) configuration of a perturbed version of the model.
Sample-Efficient L0-L2 Constrained Structure Learning of Sparse Ising Models
We consider the problem of learning the underlying graph of a sparse Ising model with $p$ nodes from $n$ i. i. d.
Query Training: Learning a Worse Model to Infer Better Marginals in Undirected Graphical Models with Hidden Variables
Here we introduce query training (QT), a mechanism to learn a PGM that is optimized for the approximate inference algorithm that will be paired with it.
Learning Sparse Classifiers: Continuous and Mixed Integer Optimization Perspectives
We aim to bridge this gap in computation times by developing new MIP-based algorithms for $\ell_0$-regularized classification.
An error bound for Lasso and Group Lasso in high dimensions
We leverage recent advances in high-dimensional statistics to derive new L2 estimation upper bounds for Lasso and Group Lasso in high-dimensions.
Improved error rates for sparse (group) learning with Lipschitz loss functions
For L1 and Slope regularizations, our bounds scale as $(k^*/n) \log(p/k^*)$ -- $n\times p$ is the size of the design matrix and $k^*$ the dimension of the theoretical loss minimizer $\B{\beta}^*$ -- and match the optimal minimax rate achieved for the least-squares case.
Learning higher-order sequential structure with cloned HMMs
We show that by constraining HMMs with a simple sparsity structure inspired by biology, we can make it learn variable order sequences efficiently.
Solving L1-regularized SVMs and related linear programs: Revisiting the effectiveness of Column and Constraint Generation
The linear Support Vector Machine (SVM) is a classic classification technique in machine learning.
Error bounds for sparse classifiers in high-dimensions
We prove an L2 recovery bound for a family of sparse estimators defined as minimizers of some empirical loss functions -- which include hinge loss and logistic loss.
Statistics Theory Statistics Theory
Hierarchical Modeling and Shrinkage for User Session Length Prediction in Media Streaming
In this work we present a novel framework inspired by hierarchical Bayesian modeling to predict, at the moment of login, the amount of time a user will spend in the streaming service.
Subset Selection with Shrinkage: Sparse Linear Modeling when the SNR is low
We conduct an extensive theoretical analysis of the predictive properties of the proposed approach and provide justification for its superior predictive performance relative to best subset selection when the noise-level is high.
