Stabilizes policy optimization with off-policy data using divergence augmentation.
problem Premature convergence and instability in policy optimization with off-policy data.
method Incorporates Bregman divergence between behavior and current policies to ensure safe policy updates.
result Empirically shows better performance in data-scarce scenarios compared to other algorithms.
RényiCL uses Rényi divergence for robust contrastive learning with stronger data augmentations.
problem Learning useful representations from multiple data views with hard augmentations.
method RényiCL employs Rényi divergence for contrastive learning, using a novel variational objective to manage hard negative sampling.
result RényiCL achieves better performance with stronger augmentations compared to other methods.
Method generates intermediate domains to align source and target domains.
problem Challenges of domain adaptation with significant domain divergence.
method Progressive domain augmentation via domain interpolation and multiple subspace alignment.
result Achieves state-of-the-art performance on multiple domain adaptation tasks.
SBA improves neural network generalization by dynamically augmenting data.
problem Lack of direct supervision in data augmentation for generalization.
method Stochastic Batch Augmentation with dynamic soft label regularization.
result SBA improves generalization and speeds up training.
Generative data augmentation boosts learning performance in various tasks.
problem Theoretical understanding of generative data augmentation's effect.
method Established a stability bound for non-i.i.d. settings, analyzed Gaussian mixture models and generative adversarial nets.
result Generative data augmentation can improve learning guarantees, especially in small train sets.
The study examines mixing times of data-augmentation Gibbs samplers for high-dimensional probit regression.
problem Investigating convergence properties of data-augmentation samplers for Bayesian probit regression.
method Using recent results on Gibbs samplers for log-concave targets, the study provides non-asymptotic bounds on mixing times.
result Explicit non-asymptotic bounds on mixing times depend on design matrix and prior precision, holding uniformly over responses.
New statistical theory explains contrastive learning effectiveness.
problem Understanding why contrastive learning works well for representation extraction.
method Developed a new theoretical framework based on approximate sufficient statistics.
result Near-sufficient encoders derived from contrastive learning can be adapted for downstream tasks.
New Wasserstein divergence improves generative model robustness and structure preservation.
problem Improving generative model robustness and structure preservation.
method Introduces a novel Wasserstein-1 path-space divergence and a WUP theorem.
result Derives robustness and generalization bounds for flow-based models.
Causal KL improves on existing metrics for evaluating causal models.
problem Insufficient discrimination between causal models using edit-distance and KL divergence.
method Introducing Causal KL, an augmented KL divergence that considers causal relationships.
result Causal KL variants effectively distinguish between observationally equivalent models.
The variational framework for learning inducing variables (Titsias, 2009a) has had a large impact on the Gaussian process literature. The framework may be interpreted as minimizing a rigorously defined Kullback-Leibler divergence between the approximating and posterior processes. To our knowledge this connection has th…
To ensure stability of learning, state-of-the-art generalized policy iteration algorithms augment the policy improvement step with a trust region constraint bounding the information loss. The size of the trust region is commonly determined by the Kullback-Leibler (KL) divergence, which not only captures the notion of d…
New method enhances adversarial robustness of deep learning models.
problem Improving the robustness of deep learning models against adversarial attacks.
method Optimal transport regularized divergences applied to distributionally robust optimization.
result Improved adversarial robustness on CIFAR-10 and CIFAR-100 datasets.
A new approach to model rejection using density ratios.
problem Improving model performance through selective prediction.
method Optimization of a loss's risk with φ-divergence regularization to find an idealized data distribution.
result Model rejection can be made by comparing the density ratio of the idealized distribution to the actual data distribution.
Entropy regularization improves policy optimization in reinforcement learning.
problem Improving policy optimization in reinforcement learning.
method Entropy regularization is introduced to soften the greedy policy towards a more diverse softmax policy, leading to a continuously parameterized algorithm that interpolates between policy gradient and Q-learning.
result An intermediate algorithm can improve performance in reinforcement learning.
Unified framework for estimating density ratios in causal inference.
problem Estimating density ratios for causal inference is challenging due to instability and curse of dimensionality.
method Bregman-Riesz regression unifies three methods: Bregman divergences, probabilistic classification, and Riesz loss.
result Unified framework improves density ratio estimation in causal inference.
Enhances crypto-asset AMM with deep learning for better liquidity and efficiency.
problem Reduced slippage and improved liquidity in decentralized finance.
method Deep reinforcement learning for predicting market equilibrium and optimizing liquidity.
result Improved capital efficiency and reduced slippage for crypto-asset traders.
The variational autoencoder (VAE) framework remains a popular option for training unsupervised generative models, especially for discrete data where generative adversarial networks (GANs) require workaround to create gradient for the generator. In our work modeling US postal addresses, we show that our discrete VAE wit…
New measure quantifies contrastive self-supervised learning's generalization ability.
problem Limited theoretical understanding of contrastive self-supervised learning's generalization.
method Defined (σ,δ)-measure to mathematically quantify data augmentation and provide an upper bound for downstream classification error. result Generalization ability is related to alignment of positive samples, divergence of class centers, and concentration of augmented data.
The family of f-divergences is ubiquitously applied to generative modeling in order to adapt the distribution of the model to that of the data. Well-definedness of f-divergences, however, requires the distributions of the data and model to overlap completely in every time step of training. As a result, as soon as the s…
New framework explains data augmentation's role in machine learning.
problem Understanding how data augmentation affects generalization and invariance learning.
method Information-theoretic framework based on mutual information bounds and orbit-averaged loss functions.
result Derives a new generalization bound decomposing the generalization gap into three interpretable terms.
Paper resolves bias in ALFT training using generalized alignment games.
problem Systematic bias in estimating logarithmic rewards from small batches.
method Generalized Distributional Alignment Games, U-statistics, minimax polynomial estimators, Variance-Optimal Augmented Polynomial Optimization Program (AQP) Estimator.
result Proves optimal bias and accelerated convergence in ALFT training.
Recent work in variational inference (VI) uses ideas from Monte Carlo estimation to tighten the lower bounds on the log-likelihood that are used as objectives. However, there is no systematic understanding of how optimizing different objectives relates to approximating the posterior distribution. Developing such a conn…
Can neural networks learn to compare graphs without feature engineering? In this paper, we show that it is possible to learn representations for graph similarity with neither domain knowledge nor supervision (i.e.\ feature engineering or labeled graphs). We propose Deep Divergence Graph Kernels, an unsupervised method …
The variational autoencoder (VAE) framework is a popular option for training unsupervised generative models, featuring ease of training and latent representation of data. The objective function of VAE does not guarantee to achieve the latter, however, and failure to do so leads to a frequent failure mode called posteri…
Paper develops tighter risk certificates for contrastive learning models.
problem Statistical theory for contrastive learning is lacking, especially for practical models like SimCLR.
method Develops non-vacuous PAC-Bayesian risk certificates considering practical SimCLR factors.
result Risk certificates for contrastive loss and downstream prediction are much tighter than previous results.
New algorithms sample convex bodies using Markov chains and restricted Gaussian oracles.
problem Sampling uniformly from convex bodies efficiently.
method Markov chain Monte Carlo with proximal sampler and restricted Gaussian oracle.
result Efficient implementation of RGO for uniform sampling on convex bodies.
We introduce several techniques for sampling and visualizing the latent spaces of generative models. Replacing linear interpolation with spherical linear interpolation prevents diverging from a model's prior distribution and produces sharper samples. J-Diagrams and MINE grids are introduced as visualizations of manifol…
Compared to supervised learning, semi-supervised learning reduces the dependence of deep learning on a large number of labeled samples. In this work, we use a small number of labeled samples and perform data augmentation on unlabeled samples to achieve image classification. Our method constrains all samples to the pred…
This work extends ME-RL using diffusion models to sample optimal policies.
problem Sampling from the optimal policy trajectory distribution in ME-RL.
method Introducing Diffusion-Augmented Markov Decision Processes (DA-MDPs) to minimize reverse KL divergence.
result DA-MDPs enable seamless integration into various ME-RL methods and outperform baselines.
Graphon autoencoder generates graphs with arbitrary sizes using Chebyshev filters.
problem Generating graphs with arbitrary sizes and arbitrary structures.
method Induces graphons from observed graphs, uses Chebyshev filters for latent representation, and learns encoder and decoder to minimize Wasserstein distance.
result Graphon autoencoder provides a new paradigm for graph generation with good generalizability and transferability.
Proposes a method to improve treatment policies in data-scarce clinical settings.
problem Improving treatment policies in data-scarce clinical settings with unobserved confounding.
method Uses a causal mechanism to model the underlying generative process and augments counterfactual trajectories with source domain priors.
result Significantly improves treatment policy performance in a simulated sepsis treatment task.
Noisy labels are ubiquitous in real-world datasets, which poses a challenge for robustly training deep neural networks (DNNs) since DNNs can easily overfit to the noisy labels. Most recent efforts have been devoted to defending noisy labels by discarding noisy samples from the training set or assigning weights to train…
We present an efficient algorithm for maximum likelihood estimation (MLE) of exponential family models, with a general parametrization of the energy function that includes neural networks. We exploit the primal-dual view of the MLE with a kinetics augmented model to obtain an estimate associated with an adversarial dua…
Novel Bayesian framework for Poisson inverse problems using Bregman geometry.
problem Solving Poisson inverse problems with non-Euclidean geometry and positivity constraints.
method Develops a Monte Carlo sampling algorithm that accounts for Bregman geometry, data augmentations, and conditional conjugacy properties.
result Efficient sampling via Gibbs steps and Hessian Riemannian Langevin Monte Carlo (HRLMC) for positivity constraints.
Improves deep learning robustness by enforcing local and global compactness.
problem Deep neural networks' vulnerability to adversarial attacks.
method Proposes Adversary Divergence Reduction Network (ADRN) that enforces local/global compactness and clustering assumption.
result Augmenting adversarial training with ADRN components improves robustness.
We consider the problem of maximum a posteriori (MAP) inference in discrete graphical models. We present a parallel MAP inference algorithm called Bethe-ADMM based on two ideas: tree-decomposition of the graph and the alternating direction method of multipliers (ADMM). However, unlike the standard ADMM, we use an inexa…
New model for community detection with side information improves recovery accuracy.
problem Community detection in networks with additional node data.
method Data Block Model (DBM) with Chernoff--TV divergence for threshold characterization and efficient algorithm.
result Sharp exact recovery threshold and efficient algorithm for DBM.
End-to-end approach learns pseudo-distance for verifying example sets.
problem Verifying example sets from same or different classes.
method Jointly trains parametric pseudo-distance with encoder.
result End-to-end approach outperforms simple scorers and classifiers.
Given two distinct datasets, an important question is if they have arisen from the the same data generating function or alternatively how their data generating functions diverge from one another. In this paper, we introduce an approach for measuring the distance between two datasets with high dimensionality using varia…
Flow AIS Bootstrap improves flow training by generating samples in hard-to-reach regions.
problem Training flows with high variance and mode-seeking behavior.
method Augment flows with AIS and minimize α-divergence with α=2. result FAB learns Boltzmann distribution of alanine dipeptide without MD samples.
In this paper, we introduce new classes of divergences by extending the definitions of the Bregman divergence and the skew Jensen divergence. These new divergence classes (g-Bregman divergence and skew g-Jensen divergence) satisfy some properties similar to the Bregman or skew Jensen divergence. We show these g-diverge…
New method controls classifier guidance in diffusion models.
problem Improving classifier guidance in diffusion models.
method Cross-entropy control of classifier gradients.
result Effective guidance vectors with mean squared error O(dε). Federated learning (FL) is a distributed deep learning method which enables multiple participants, such as mobile phones and IoT devices, to contribute a neural network model while their private training data remains in local devices. This distributed approach is promising in the edge computing system where have a larg…
Variational Bayes (VB), also known as independent mean-field approximation, has become a popular method for Bayesian network inference in recent years. Its application is vast, e.g. in neural network, compressed sensing, clustering, etc. to name just a few. In this paper, the independence constraint in VB will be relax…
Divergence functions play a key role as to measure the discrepancy between two points in the field of machine learning, statistics and signal processing. Well-known divergences are the Bregman divergences, the Jensen divergences and the f-divergences. In this paper, we show that the symmetric Bregman divergence can be …
Stabilizes GAN training with limited data.
problem Overfitting in GANs with scarce data.
method Adaptive discriminator augmentation.
result Good results possible with few thousand images.
A new recursive mixture estimation algorithm improves VAE inference efficiency and accuracy.
problem Inaccurate posterior approximation in traditional VAEs.
method Recursive mixture estimation algorithm using functional gradient approach for iterative component selection.
result Significantly higher test data likelihood compared to state-of-the-art methods on benchmark datasets.
Study explores relationship between Hölder and FDPD divergences.
problem Understanding the relationship between Hölder and FDPD divergences.
method Intersection and generalization of divergence families, proving nonnegativity, deriving inequalities.
result Established ξ-Hölder divergence and derived inequalities.