BONAS accelerates NAS while maintaining reliability.
problem Computational inefficiency in sample-based NAS.
method Bayesian Optimized Neural Architecture Search (BONAS) using weight-sharing.
result BONAS accelerates sample-based NAS significantly while maintaining reliability.
The study proves sampling-based GNNs can approximate training on full graphs with small subgraphs.
problem Training Graph Neural Networks (GNNs) on large graphs is computationally expensive.
method Theoretical framework using graph local limits to prove approximation of GNN training on small samples.
result Parameters learned from sampling-based GNNs on small subgraphs are close to those on full graphs.
New method for Bayesian neural networks reduces inference difficulty.
problem Difficulty in sample-based inference for Bayesian neural networks.
method Embracing mode-connectedness to link overparameterization and sampling difficulty.
result Practical guidelines and deep ensemble approach for effective SBI.
This paper presents sampling-based speech parameter generation using moment-matching networks for Deep Neural Network (DNN)-based speech synthesis. Although people never produce exactly the same speech even if we try to express the same linguistic and para-linguistic information, typical statistical speech synthesis pr…
Posterior sampling-based EI achieves sublinear regret bounds for expensive function optimization.
problem Theoretical analysis of expected improvement (EI) in Bayesian optimization.
method Randomized posterior sampling of EI.
result Achieves sublinear Bayesian cumulative regret bounds.
Given a data matrix X∈Rn×d and a response vector y∈Rn, suppose n>d, it costs O(nd2) time and O(nd) space to solve the least squares regression (LSR) problem. When n and d are both large, exactly solving the LSR problem is very expensive. When n≫d, one feasible approach to spee…
Improves generative models for cost-sensitive decisions.
problem Generative models lack awareness of decision costs.
method Integrates a decision loss into the training objective.
result Improves cost-sensitive forecast accuracy.
Quantum algorithm improves portfolio construction accuracy.
problem Efficiently constructing portfolios with real-world constraints.
method Sampling-based CVaR Variational Quantum Algorithm (VQA) combined with local-search post-processing.
result Achieved a relative solution error of 0.49% on IBM Heron processors.
A scalable method for Bayesian inference in large linear models.
problem High computational cost in Bayesian linear models for large networks.
method Sample-based inference and g-prior for hyperparameter selection.
result Linearised neural network inference on large datasets (ResNet-18, ResNet-50, U-Net).
We study primal-dual type stochastic optimization algorithms with non-uniform sampling. Our main theoretical contribution in this paper is to present a convergence analysis of Stochastic Primal Dual Coordinate (SPDC) Method with arbitrary sampling. Based on this theoretical framework, we propose Optimality Violation-ba…
We present an embedding of stochastic optimal control problems, of the so called path integral form, into reproducing kernel Hilbert spaces. Using consistent, sample based estimates of the embedding leads to a model free, non-parametric approach for calculation of an approximate solution to the control problem. This fo…
Improves generalization with few samples using a new regularization method.
problem Training deep neural networks with limited data leads to overfitting.
method Sample-based regularization (SBR) to improve generalization without relying on source model knowledge.
result SBR outperformed existing methods in various configurations.
This work compares and evaluates various sampling methods for neural language models.
problem Lack of systematic comparison and myths about sampling methods.
method Monte Carlo sampling, importance sampling, compensated partial summation, noise contrastive estimation.
result All sampling methods can perform equally well if posterior probabilities are corrected.
A new method improves SVI for high-dimensional, poorly-conditioned distributions.
problem Challenges in existing SVI methods for high-dimensional, poorly-conditioned distributions.
method Trust-region optimization approach leveraging conditional independences and second-order information.
result Superior numerical performance and better scalability in high-dimensional distributions.
We study the problem of finding the optimal dosage in early stage clinical trials through the multi-armed bandit lens. We advocate the use of the Thompson Sampling principle, a flexible algorithm that can accommodate different types of monotonicity assumptions on the toxicity and efficacy of the doses. For the simplest…
Deep unfolding accelerates MCMC-based COP solvers.
problem Optimizing combinatorial problems with MCMC and gradient descent.
method Combines MCMC and gradient descent, trains step sizes, uses variance estimation for non-differentiable MCMC.
result Significantly accelerates convergence speed for COPs.
Bayesian model-based reinforcement learning is a formally elegant approach to learning optimal behaviour under model uncertainty, trading off exploration and exploitation in an ideal way. Unfortunately, finding the resulting Bayes-optimal policies is notoriously taxing, since the search space becomes enormous. In this …
New sampling-based approach for filtering problems using multiplicative Gaussian functions.
problem Approximate inference in filtering problems.
method Approximates distribution with a weighted sum of continuous functions using sampling for multiplications.
result Preliminary experiments show potential of the new method compared to particle filters.
Domain adaptation has become a prominent problem setting in machine learning and related fields. This review asks the question: how can a classifier learn from a source domain and generalize to a target domain? We present a categorization of approaches, divided into, what we refer to as, sample-based, feature-based and…
Motion planning for robots of high degrees-of-freedom (DOFs) is an important problem in robotics with sampling-based methods in configuration space C as one popular solution. Recently, machine learning methods have been introduced into sampling-based motion planning methods, which train a classifier to distinguish coll…
Differentiable learning via SGD and GD can simulate various learning problems, depending on precision and minibatch size.
problem Understanding the power of differentiable learning via SGD and GD compared to statistical query (SQ) learning.
method Comparing the learning power of SGD and GD on population and empirical losses with statistical query learning.
result The learning power of SGD and GD depends on the precision of gradient calculations relative to the minibatch size or sample size.
MIRA scores assess conditional distribution accuracy using joint samples.
problem Assessing the accuracy of candidate conditional distributions.
method Analytic expression for Mira score based on equal probability mass regions.
result Mira enables Bayesian model comparison by quantifying alignment with true process.
Efficiently certifies global robustness of large neural networks with probabilistic guarantees.
problem Certifying robustness of large neural networks in a scalable and efficient manner.
method Sampling an ε-net and invoking a local robustness oracle.
result Certifies a probabilistic relaxation of robustness efficiently and globally.
Bounded rationality investigates utility-optimizing decision-makers with limited information-processing power. In particular, information theoretic bounded rationality models formalize resource constraints abstractly in terms of relative Shannon information, namely the Kullback-Leibler Divergence between the agents' pr…
Efficient clustering for large datasets using a sampling-based approach.
problem Clustering high-dimensional data with a large number of clusters efficiently.
method A simple and efficient clustering method that evaluates distances of data points with a subset of cluster centers.
result Optimal solutions of the approximation are the same as in the exact solution, but more efficient at extracting clusters.
Efficient algorithm finds fast Transformer models.
problem Slow inference time of Transformer models.
method Decompose Transformer architecture into components, use sampling-based one-shot search.
result Achieved 10% to 30% speedup on pre-trained BERT and 70% on top of a previous state-of-the-art model.
A framework for hypothesis testing on attributed graphs using sampling.
problem Statistical testing on graph data, especially large attributed graphs.
method Sampling-based framework with PHASE and PHASEopt for accurate and efficient hypothesis testing.
result PHASE and PHASEopt improve accuracy and efficiency of hypothesis testing in attributed graphs.
Evaluating generative adversarial networks (GANs) is inherently challenging. In this paper, we revisit several representative sample-based evaluation metrics for GANs, and address the problem of how to evaluate the evaluation metrics. We start with a few necessary conditions for metrics to produce meaningful scores, su…
HH-VAEM improves imputation and acquisition of missing data using hierarchical models and Hamiltonian Monte Carlo.
problem Imputation and acquisition of missing heterogeneous data.
method Hierarchical VAE model with Hamiltonian Monte Carlo and automatic hyper-parameter tuning.
result HH-VAEM outperforms existing methods in imputation and supervised learning tasks.
We propose a relaxation-based approximate inference algorithm that samples near-MAP configurations of a binary pairwise Markov random field. We experiment on MAP inference tasks in several restricted Boltzmann machines. We also use our underlying sampler to estimate the log-partition function of restricted Boltzmann ma…
To meet the standard of differential privacy, noise is usually added into the original data, which inevitably deteriorates the predicting performance of subsequent learning algorithms. In this paper, motivated by the success of improving predicting performance by ensemble learning, we propose to enhance privacy-preserv…
Graph Convolutional Networks (GCNs) are powerful models for learning representations of attributed graphs. To scale GCNs to large graphs, state-of-the-art methods use various layer sampling techniques to alleviate the "neighbor explosion" problem during minibatch training. We propose GraphSAINT, a graph sampling based …
Neural solver computes Wasserstein geodesics and velocity fields efficiently.
problem Computing Wasserstein geodesics and velocity fields efficiently.
method Sample-based neural network approach to solve the minimax problem.
result Directly samples from target distribution and estimates velocity field.
PS4POMDPs algorithm simplifies online learning for episodic POMDPs with unknown models.
problem Learning in POMDPs is harder than in MDPs; online learning is especially challenging.
method Posterior Sampling-based reinforcement learning algorithm (PS4POMDPs)
result Bayesian regret scales as √number of episodes and is polynomial in other parameters.
Improved Thompson Sampling algorithms for bandits with tighter regret bounds.
problem Efficient and adaptive algorithms for stochastic bandits with bounded rewards.
method Proposed two parameterized Thompson Sampling-based algorithms: TS-MA-α and TS-TD-α.
result Achieved O(Kln^(α+1)(T)/Δ) regret bound, improving scalability and resource allocation.
New method learns graph structure and uncertainty from data.
problem Learning latent graph structures and their uncertainty from data.
method Proposes a sampling-based method to learn latent graph structure and uncertainty simultaneously.
result Proves that suitable loss functions on stochastic model outputs solve both learning latent graph structure and achieving optimal predictions.
DGFS improves sampling from complex densities by optimizing partial trajectories.
problem Sampling from intractable high-dimensional density functions.
method DGFS uses a flow function to break down the training process into short partial trajectory segments, leveraging intermediate learning signals.
result DGFS achieves more accurate estimates of the normalization constant.
SDPG algorithm improves sample efficiency and reward in DRL for continuous action spaces.
problem Improving sample efficiency and reward in distributional reinforcement learning for continuous action spaces.
method SDPG algorithm models return distribution using samples via reparameterization technique.
result SDPG shows better sample efficiency and higher reward in OpenAI Gym environments.
Paper proposes robust methods for estimating optimal treatment rules with censored survival data.
problem Estimating optimal treatment rules for censored survival data.
method Developed two robust criteria and a sampling-based difference-of-convex algorithm for learning optimal treatment rules.
result Proposed methods show improved performance compared to existing methods in simulations and real data.
We provide initial seedings to the Quick Shift clustering algorithm, which approximate the locally high-density regions of the data. Such seedings act as more stable and expressive cluster-cores than the singleton modes found by Quick Shift. We establish statistical consistency guarantees for this modification. We then…
New method for efficient marginalization of discrete latent variables in neural networks.
problem Computational challenges in training models with discrete latent variables.
method Parameterizing discrete distributions using sparse mappings (sparsemax and structured variants) to reduce support and enable efficient marginalization.
result Achieved good performance in various tasks with efficient and practical training.
LaPSRL achieves optimal regret for isoperimetric RL distributions.
problem Designing RL algorithms with sublinear regret for non-log-concave distributions.
method Posterior Sampling (PSRL) and Langevin sampling (LaPSRL) for isoperimetric distributions.
result LaPSRL achieves order-optimal regret and subquadratic complexity.
Paper proposes first unlearning algorithm for MCMC models.
problem Enforcing right to be forgotten in AI causes high costs for data deletion.
method Converts MCMC unlearning to explicit optimization problem, designs MCMC influence function.
result MCMC unlearning does not compromise generalizability of models.
Purpose: Machine learning is broadly used for clinical data analysis. Before training a model, a machine learning algorithm must be selected. Also, the values of one or more model parameters termed hyper-parameters must be set. Selecting algorithms and hyper-parameter values requires advanced machine learning knowledge…
SBBO optimizes complex spaces using sampling-based models.
problem Optimizing complex spaces with discrete variables.
method Simulation Based Bayesian Optimization (SBBO) using sampling-based surrogate models.
result Empirical effectiveness of SBBO in combinatorial optimization.
Low-rank matrix approximations are often used to help scale standard machine learning algorithms to large-scale problems. Recently, matrix coherence has been used to characterize the ability to extract global information from a subset of matrix entries in the context of these low-rank approximations and other sampling-…
This guide explains statistical distances for evaluating generative models.
problem Evaluating the quality of samples from generative models.
method Four statistical distances: SW, C2ST, MMD, FID.
result Different distances can yield varying results on similar data.
LLMs generate answers under incomplete context, and their uncertainty should scale with missing information.
problem Evaluating the quality of LLM answers under incomplete context.
method A controlled framework with varying context availability, and two uncertainty measures (sampling-based confidence and response entropy) evaluated on SQuAD.
result Response entropy increases with context removal and explains more variance in accuracy than confidence, suggesting it is a more responsive uncertainty measure.