arXiv research

Unified framework for tractable inference scenarios in machine learning models.

Probabilistic models learned as density estimators can be exploited in representation learning beside being toolboxes used to answer inference queries only. However, how to extract useful representations highly depends on the particular model involved. We argue that tractable inference, i.e. inference that can be compu…

TRUST improves structure learning with tractable uncertainty.

New tractable density models from squaring neural networks.

Normalizing flows can now estimate densities on unknown manifolds.

A central problem in machine learning involves modeling complex data-sets using highly flexible families of probability distributions in which learning, sampling, inference, and evaluation are still analytically or computationally tractable. Here, we develop an approach that simultaneously achieves both flexibility and…

New model generates data on constrained sets without losing tractability.

Study shows tractable generalization in RL is impossible but possible with Strong Proximity.

Normalizing Flows model tractable distributions for efficient sampling and evaluation.

Researchers compare lower bounds and tractability in reinforcement learning.

We develop computationally efficient Riemannian manifolds for graph embeddings.

In this paper we present decomposable priors, a family of priors over structure and parameters of tree belief nets for which Bayesian learning with complete observations is tractable, in the sense that the posterior is also decomposable and can be completely determined analytically in polynomial time. This follows from…

The paper explores how to learn models that respect constraints in probabilistic learning.

ACVAEs improve on CVAEs by learning more flexible latent correlations.

A new heuristic for learning Markov network structure efficiently.

PNCs balance tractability and expressiveness in probabilistic modeling.

New framework integrates SPNs with weighted model integration for hybrid data.

Unified tractability conditions for various compositional inference queries.

CCs learn high-dimensional distributions from heterogeneous data.

New algorithm achieves optimal regret in average reward MDPs without prior bias information.

New perspective on distribution shift helps make learning easier.

Paper shows how to infer hidden states in neural networks analytically.

Casting neural networks in generative frameworks is a highly sought-after endeavor these days. Contemporary methods, such as Generative Adversarial Networks, capture some of the generative capabilities, but not all. In particular, they lack the ability of tractable marginalization, and thus are not suitable for many ta…

This work tackles slate-based recommender systems using RL, optimizing long-term user engagement.

We study the computational tractability of PAC reinforcement learning with rich observations. We present new provably sample-efficient algorithms for environments with deterministic hidden state dynamics and stochastic rich observations. These methods operate in an oracle model of computation -- accessing policy and va…

DBKs enable scalable GPs with tractable inference for large datasets.

Study efficient interactive learning for structured outputs with reliable computation.

XSPNs combine SPNs and MEVMs for efficient inference in data with repeated parts.

Hybrid model combines continuous and tractable probabilistic models.

TRAK traces model predictions to training data efficiently.

We consider the problem of transforming samples from one continuous source distribution into samples from another target distribution. We demonstrate with optimal transport theory that when the source distribution can be easily sampled from and the target distribution is log-concave, this can be tractably solved with c…

Despite the fundamental nature of the inhomogeneous Poisson process in the theory and application of stochastic processes, and its attractive generalizations (e.g. Cox process), few tractable nonparametric modeling approaches of intensity functions exist, especially when observed points lie in a high-dimensional space.…

A new algorithm approximates optimal stopping problems with semi-tractable complexity.

Improved video tracking accuracy with active learning.

We introduce RNADE, a new model for joint density estimation of real-valued vectors. Our model calculates the density of a datapoint as the product of one-dimensional conditionals modeled using mixture density networks with shared parameters. RNADE learns a distributed representation of the data, while having a tractab…

Information theoretic active learning has been widely studied for probabilistic models. For simple regression an optimal myopic policy is easily tractable. However, for other tasks and with more complex models, such as classification with nonparametric models, the optimal solution is harder to compute. Current approach…

New algorithms for efficient inference and sampling in complex Ising models.

Proposes SQUAD for better predictive uncertainty in deep latent models.

Fixed-parameter tractability of private synthetic data generation

Simplifies inference for simulators with or without tractable likelihoods.

We present a novel tractable generative model that extends Sum-Product Networks (SPNs) and significantly boosts their power. We call it Sum-Product-Quotient Networks (SPQNs), whose core concept is to incorporate conditional distributions into the model by direct computation using quotient nodes, e.g. $P(A|B) = \frac{P(…

Paper tackles expected predictions computation for arbitrary generative models.

Proposes Exogenous Matching for efficient counterfactual estimation.

Proves efficient learning of hierarchical structure in meta-reinforcement learning.

Many polynomial invariants of knots and links, including the Jones and HOMFLY-PT polynomials, are widely used in practice but #P-hard to compute. It was shown by Makowsky in 2001 that computing the Jones polynomial is fixed-parameter tractable in the treewidth of the link diagram, but the parameterised complexity of th…

Deep SCMs with deep learning infer counterfactuals from noisy data.

Despite widespread interest and practical use, the theoretical properties of random forests are still not well understood. In this paper we contribute to this understanding in two ways. We present a new theoretically tractable variant of random regression forests and prove that our algorithm is consistent. We also prov…

New method approximates complex kernel norms with random features, making learning tractable.