New method reduces the gap between dropout training and inference phases.
problem The gap between dropout training and inference phases in neural networks.
method Formulate dropout as a tractable approximation of a latent variable model, introduce expectation-linear dropout neural networks, and regularize the training objective.
result Reduction in the inference gap can improve performance consistently.
New method uses model's generalization gap to predict membership inference attacks.
problem Predicting membership inference attacks on machine learning models.
method Uses the model's generalization gap as a metric to estimate the vulnerability to membership inference attacks.
result Demonstrates that the model's generalization gap provides an upper bound on MIA security.
New method improves image compression using bits-back coding.
problem Lossy image compression with deep latent variable models.
method Iterative inference, stochastic annealing, bits-back coding.
result New state-of-the-art performance on lossy image compression.
Two methods for fast Gaussian process inference on large datasets with gaps.
problem Efficient Gaussian process inference for large datasets with missing values.
method Two novel approaches using Kronecker matrix algebra and low-rank preconditioning.
result The second method is greatly advantageous for problems with gaps.
Study shows how to reduce variational inference bias by concentrating likelihood ratio distribution.
problem Bias and variance issues in variational inference.
method Upper bound variational gap using dispersion measure of likelihood ratio, suggesting methods to reduce bias.
result Reducing bias in variational inference can be achieved by making likelihood ratio distribution more concentrated.
Improves VAE training by refining variational parameters with BSVI.
problem Amortized inference in VAEs leads to suboptimal variational parameters and the amortization gap.
method Proposes BSVI, a refinement procedure using SVI's importance weights.
result Training VAEs with BSVI yields improved performance compared to SVI.
Iterative models improve inference efficiency in deep latent variable models.
problem Inference models in deep latent variable models are computationally inefficient and have an amortization gap.
method Proposes iterative models that learn to perform inference optimization through repeated encoding of gradients.
result Iterative models outperform standard inference models on benchmark data sets of images and text.
Recent advances in stochastic gradient variational inference have made it possible to perform variational Bayesian inference with posterior approximations containing auxiliary random variables. This enables us to explore a new synthesis of variational inference and Monte Carlo methods where we incorporate one or more s…
New method improves generalization of VAEs by reducing overfitting.
problem Generalization issues in VAEs overfitting to training data.
method Proposed a new training objective to improve amortized inference.
result Improved performance in image modeling and lossless compression.
Efficient method for tensor linear form inference with noisy incomplete data.
problem Statistical inference of tensor linear forms with incomplete and noisy observations.
method Initial estimate + debiasing + one-step power iteration.
result Optimal uncertainty quantification and statistical-to-computational gaps examined.
Paper tackles small eigen-gap estimation and inference for noisy symmetric matrices.
problem Estimating eigenvectors with small eigen-gap and fine-grained statistical reasoning.
method Eigen-decomposition of asymmetric data matrix, distribution-free procedures, adaptive to heteroscedastic noise.
result Minimax optimal under Gaussian noise, confidence intervals for eigenvalues, small eigen-gap handling.
A new method simulates a lazy version of a Markov chain for empirical inference.
problem Estimating and testing unknown Markov chains with limited data.
method Simulates an α-lazy version of an unknown Markov chain, making it ergodic.
result The pseudo spectral gap can be applied to non-ergodic Markov chains.
The paper proposes a method for reliable statistical inference on OSNs using random walks and Bayesian techniques.
problem Limited API crawls pose challenges for reliable statistical inference in online social networks.
method The paper uses regenerative properties of random walks to propose an unbiased estimator and derives an approximate posterior distribution for Bayesian inference.
result The proposed method provides a way to make reliable statistical inferences with limited API crawls.
Detecting and recovering labels in binomial logistic mixtures is challenging due to an information gap.
problem Detecting and recovering labels in binomial logistic mixtures
method Propose two feasibility-aware inference procedures
result Avoid misleading component selections and improve label probability calibration
New Holder bounds improve variational inference by flattening thermodynamic curves.
problem Improving variational inference by addressing performance gaps between theory and practice.
method Generalizing thermodynamic integration to weighted Holder mean, introducing Holder bounds.
result Holder bounds promise a one-step approximation of exact marginal log-likelihood.
The paper addresses the gap between theoretical and practical confidence set widths in universal inference.
problem Inference procedures can be overly conservative, leading to wider confidence sets than expected.
method The authors identify the source of asymptotic conservativeness and propose a remedy based on studentization and bias correction.
result The proposed method achieves exact asymptotic coverage at the nominal 1−α level, even under model misspecification. A-VI can approximate F-VI under certain conditions, improving inference in some models.
problem Approximate Bayesian inference using factorized distributions.
method Amortized variational inference (A-VI) learns a common inference function for latent variables.
result A-VI can close the amortization gap in simple hierarchical models.
New bounds on machine learning data leakage identified.
problem Machine Learning models can leak sensitive information.
method Formalized attack setups, derived universal bounds, studied mutual information.
result Connected attack success rate to generalization gap and mutual information.
Unified framework for causal inference under sample selection.
problem Causal inference under sample selection with treatment and outcome non-randomness.
method ForestRiesz estimator, Riesz representation framework.
result ForestRiesz estimator yields more stable treatment effect estimates than conventional double machine learning approaches.
This paper bridges variational inference and Wasserstein gradient flows.
problem Combining variational inference and Wasserstein gradient flows for more efficient approximations.
method Recasting Bures-Wasserstein gradient flow as a Euclidean gradient flow and using path-derivative gradient estimator.
result A new gradient estimator for f-divergences that can be implemented using machine learning libraries. CausalML simplifies causal inference methods in Python.
problem Combining causal inference and machine learning.
method Collection of causal inference methods in Python.
result Makes causal inference methods accessible in Python.
New algorithm identifies missing data models using directed acyclic graphs.
problem Identifying missing data distributions represented by DAGs.
method Proposed a new algorithm that generalizes ID algorithm for causal inference.
result Significantly generalizes types of manipulations for identification.
New method explains computational barriers in high-dimensional statistical models.
problem Understanding detection-recovery gaps in high-dimensional inference.
method Combining algorithmic contiguity and cross-validation reduction to obtain conditional computational lower bounds.
result Mild control of low-degree advantage is sufficient to explain computational barriers for recovery.
In this tutorial we explain the inference procedures developed for the sparse Gaussian process (GP) regression and Gaussian process latent variable model (GPLVM). Due to page limit the derivation given in Titsias (2009) and Titsias & Lawrence (2010) is brief, hence getting a full picture of it requires collecting resul…
Improves Bayesian predictive performance in misspecified models.
problem Misspecification gap between inferential and predictive risks.
method Develops a multi-sample loss (PACm) to bridge the gap. result Empirical study shows improved predictive distribution.
Simulation-free VI closes the approximation gap in latent SDEs
problem Recovering dynamical systems from noisy observations
method Helmholtz-SDE
result Recovers dynamics more faithfully than prior methods
Improves spatio-temporal forecasting by reducing errors between training and inference.
problem Accumulation of small errors in Seq2Seq models during inference due to different distributions of training and inference phases.
method Curriculum learning based on Temporal Progressive Growing Sampling to replace some ground-truth context with generated predictions.
result Better models long-term dependencies and outperforms baseline approaches on two datasets.
GAP uses deep learning to efficiently partition graphs.
problem Graph partitioning to minimize edge cut.
method Deep learning approach with a differentiable loss function.
result GAP achieves competitive partitions and generalizes to unseen graphs.
Paper proposes efficient inference for hidden Markov models with memory decay.
problem Challenges in scalability due to dependencies in hidden Markov model observation data.
method Utilizes memory decay to carry out forward and backward probabilities with subsequences, enabling efficient inference over long sequences.
result Developed an efficient algorithm to numerically estimate the gap of top Lyapunov exponents, which determines the length of subsequences.
Polynomial-time algorithm finds planted hypercube vectors in Gaussian mixtures.
problem Clustering d-dimensional Gaussian mixtures with unknown covariance.
method Lattice-based methods using Lenstra--Lenstra--Lovasz reduction.
result Achieves statistically-optimal sample complexity of d+1 samples.
This paper improves meta-learning by developing new PAC-Bayes bounds.
problem Meta-learning generalization gap across multiple tasks.
method Upper bounding convex functions linking environment and task-level losses.
result New PAC-Bayes bounds for meta-learning with improved algorithms.
New guarantees for black-box variational inference methods.
problem Insufficient theoretical guarantees for black-box variational inference.
method Novel convergence guarantees for stochastic optimization of variational inference.
result Provable convergence of proximal and projected stochastic gradient descent for variational inference.
Sparse attention model reduces long-context inference time with exponential accuracy guarantees.
problem Efficiently processing long-context queries in large language models.
method Formalizes attention as a projection onto key vectors, analyzes entropic relaxation, and introduces Vashista Sparse Attention.
result Sparse attention concentrates on a constant-size active face, leading to exponential decay of inactive tokens' mass and linear scaling of active face error.
New insights into variational inference using Monte Carlo estimates.
problem Improving variational bounds in latent variable models.
method Analyzing properties of Monte Carlo estimates and their impact on variational gaps.
result Negative correlation reduces variational gaps, contrary to intuition.
Improved inference-time alignment using Best-of-N and smoothing.
problem Reward overoptimization in Best-of-N (BoN) due to poor proxy reward models.
method Introduced Soft Best-of-N (SBoN) and analyzed its performance through KL divergence and regret analysis.
result Smoothing helps SBoN mitigate reward overoptimization, especially when proxy reward quality is low.
New insights link diverse statistical problems via secret leakage planted clique.
problem Statistical-computational gaps in inference problems.
method Secret leakage planted clique as a new hardness assumption for reductions.
result Establishes tight statistical-computational tradeoffs for various problems.
We simplify inference for TPP models with latent structures.
problem Intractable marginalization in TPP models with latent structures.
method Approximate inference over latent variables using a tight upper bound on the approximation gap.
result Improved results for models like Survival Analysis.
Study shows computational and statistical gaps in Gaussian Single-Index Models.
problem Statistical and computational trade-offs in high-dimensional regression problems.
method Analysis of SQ and LDP frameworks, partial-trace algorithm.
result Computational algorithms require significantly more samples than information-theoretic limits.
Boosting VI improves inference by greedily combining densities.
problem Improving tractability in Bayesian statistics.
method Adapting boosting techniques to VI, replacing a single density with a mixture of densities.
result Demonstrates convergence of boosting VI under relaxed smoothness assumptions.
The paper proposes a method to test properties of the optimal assortment in multinomial logit models.
problem Uncertainty quantification for the optimal assortment in multinomial logit models.
method The paper proposes a novel inferential framework to test properties of the optimal assortment in multinomial logit models, reducing the problem to detecting the sign change point of marginal revenue gaps.
result The asymptotic normality of the marginal revenue gap estimator and the construction of a maximum statistic to detect the sign change point.
New method explains ML performance gaps without causal knowledge.
problem Understanding why ML algorithms perform differently across domains.
method Nonparametric hierarchical decomposition framework.
result Detailed variable-level explanations for performance gaps.
Real-time scene understanding solved using Approximate Bayesian Computation.
problem Predicting human actions, object poses, and pedestrian crossings from depth images.
method Bayesian error model, neural surrogates, and adaptive discretization.
result Real-time inference on real-world problems is feasible.
Unified framework to bridge human and LLM judgments.
problem Systematic discrepancies between human and LLM evaluations.
method Latent human preference score and linear transformations of covariates.
result Higher agreement with human ratings and exposure of systematic gaps.
Theoretical study explains why randomization improves adversarial robustness.
problem Lack of theoretical arguments for randomization techniques in adversarial robustness.
method Theoretical analysis of randomization techniques for adversarial robustness.
result New theoretical upper bound on adversarial generalization gap.
Boosting variational inference uses optimization to improve posterior approximations.
problem Improving the tractability of Bayesian posterior approximations.
method An optimization-based approach to greedily add components to a mixture of densities.
result Established convergence properties and rates for the boosting variational inference algorithm.
Improves model robustness to shifts in subpopulations.
problem Poor performance of ML models under data distribution shifts.
method Develops group-aware priors (GAP) over neural network parameters.
result Training with GAP yields state-of-the-art performance.
This paper analyzes speculative decoding, a method to speed up large language model inferences.
problem Theoretical understanding of speculative decoding is lacking.
method Conceptualizes speculative decoding as a markov chain problem and studies its key properties.
result Reveals fundamental connections between LLM components and their impact on decoding efficiency.
Paper combines deterministic and stochastic inference methods for PGMs.
problem Combining biases from deterministic methods and high costs from Monte Carlo.
method Sequential Monte Carlo algorithm that uses output from deterministic approximations.
result Improves upon deterministic methods and Monte Carlo by reducing biases and computational costs.