Proposes a new loss function for deep neural networks.
problem Deep neural networks lack a direct method to discriminate between correct and competing classes.
method Introduces a discriminative loss function based on negative log likelihood ratio.
result Significantly outperforms cross-entropy loss on image classification tasks.
Novel neural likelihood ratio estimation for negative data in particle physics.
problem Estimating likelihood ratios with negative probability densities and weights.
method Introducing a novel loss function and a new model architecture based on signed mixture models.
result Demonstrated improved estimation on a real-world example from particle physics.
Deep learning compresses and quantizes log-likelihood ratios for fading channels.
problem Efficiently compress and quantize log-likelihood ratios for fading channels.
method Trains a deep autoencoder network to map log-likelihood ratios to a latent space and reconstruct them.
result Achieves a compression factor of nearly three times with minimal performance loss.
wd1 improves reasoning in dLLMs by optimizing policies without policy ratios.
problem Improving reasoning in diffusion-based large language models through RL.
method wd1: ratio-free policy optimization using weighted log-likelihood.
result wd1 outperforms diffusion-based GRPO while requiring lower computational cost.
Study detects signals in spiked Wigner models using log likelihood ratio.
problem Detecting signals in rank-one spiked Wigner models with non-Gaussian noise.
method Proved asymptotic normality of log likelihood ratio and computed error thresholds.
result Optimal signal-to-noise ratio threshold for reliable detection.
New method for robust distribution alignment using log-likelihood ratio and normalizing flows.
problem Distribution alignment challenges in deep learning.
method Log-likelihood ratio statistic and normalizing flows.
result Minimizing the proposed objective yields robust domain alignment.
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.
The study uses information theory to set lower bounds on model likelihoods.
problem Improving latent variable models by optimizing priors or likelihoods.
method Applying rate-distortion theory to find lower bounds on negative log likelihood.
result Rate-distortion theory can be used to optimize priors and likelihoods in latent variable models.
PACMAN provides bounds for classification tasks considering accuracy vs. negative log-loss mismatch.
problem Mismatch between accuracy and negative log-loss in classification tasks.
method Point-wise PAC approach over generalization gap, using likelihood ratio and concentration inequalities.
result PACMAN provides point-wise PAC bounds for the generalization problem.
Maximum likelihood training improves the performance of score-based diffusion models.
problem Training score-based diffusion models with maximum likelihood.
method Trained by minimizing a weighted combination of score matching losses, with a specific weighting scheme that bounds negative log-likelihood.
result Maximum likelihood training improves the log-likelihood of score-based diffusion models across multiple datasets.
SPRT-TANDEM improves sequential classification accuracy with fewer samples.
problem Efficiently classifying sequential data with high accuracy and low sampling cost.
method Deep neural network-based SPRT algorithm that estimates log-likelihood ratio of two hypotheses.
result SPRT-TANDEM achieves statistically significantly better classification accuracy than other classifiers with fewer samples.
Aligns databases with Gaussian features using MAP estimation and thresholding.
problem Aligning databases with jointly Gaussian features.
method Two algorithms: MAP estimation and thresholding of log likelihood ratios.
result Conditions on mutual information for reliable performance of algorithms.
A new algorithm detects changes in data with constant cost per iteration.
problem Detecting changes in data with low computational cost.
method Adapting pruning and maximisation techniques from Gaussian data to exponential family models.
result The algorithm can detect changes in a wide range of models with a constant per-iteration cost.
Noise-Contrastive Estimation improves efficiency for estimating log-likelihood of complex point processes.
problem Estimating log-likelihood of complex multivariate point processes is computationally expensive.
method Noise-Contrastive Estimation adapted for multivariate point processes, with provable guarantees.
result Our method achieves similar log-likelihood with fewer evaluations and less time.
Enhances VAEs for sharper image synthesis.
problem Blurriness in generated images from VAEs.
method Integrates a downscaled version of the original image into the VAE framework and uses it as input to the decoder.
result Improves FID score in image synthesis while maintaining similar log-likelihood performance.
We exhibit a strong link between frequentist PAC-Bayesian risk bounds and the Bayesian marginal likelihood. That is, for the negative log-likelihood loss function, we show that the minimization of PAC-Bayesian generalization risk bounds maximizes the Bayesian marginal likelihood. This provides an alternative explanatio…
New methods improve anomaly detection in deep networks by leveraging hierarchical likelihoods and multi-scale features.
problem Challenges in detecting anomalies in high-level features due to model bias and domain prior.
method Two methods: 1) Log likelihood ratios between in-distribution and general distribution models, 2) Multi-scale likelihood contribution.
result Strong anomaly detection performance in unsupervised setting, slightly underperforming supervised methods.
Separable losses are inconsistent for structured prediction models.
problem Inconsistency of separable losses in structured prediction models.
method Analysis of separable negative log-likelihood losses for structured prediction.
result Separable losses are not Bayes consistent and may not predict the most probable structure.
Optimally tackles covariate shift in RKHS-based nonparametric regression.
problem Covariate shift in nonparametric regression over RKHS.
method Two families of covariate shift problems defined using likelihood ratios. Minimax rate-optimal estimators for KRR and reweighted KRR.
result KRR is minimax rate-optimal and strictly sub-optimal compared to naive estimator under covariate shift.
Extends likelihood ratio exponential families to analyze various optimization methods.
problem Analyzing optimization methods like rate-distortion and information bottleneck.
method Linking geometric mixture paths to exponential families and using hypothesis testing.
result Provides a common mathematical framework for understanding these methods.
Improved estimation for imbalanced data using log odds correction and optimal sampling.
problem Parameter estimation with nonuniform negative sampling for imbalanced data.
method Derive asymptotic distribution of IPW estimator, derive optimal sampling probability, propose likelihood-based estimator.
result Improved estimator has the smallest asymptotic variance.
Anytime-valid confirmation of label-shift corrections
problem Small-batch scientific deployments with scarce labeled outcomes
method Conditional e-value and martingale-based rule
result Nonnegative martingale and anytime-valid confirmation rule
Smart Bayes integrates generative and discriminative features for improved classification.
problem Improving classification performance by combining generative and discriminative modeling.
method Integrates generative likelihood-ratio features into a logistic-regression-style classifier.
result Often outperforms logistic regression and Naive Bayes in simulations and real data.
Researchers have constantly asked whether stock returns can be predicted by some macroeconomic data. However, it is known that macroeconomic data may exhibit nonstationarity and/or heavy tails, which complicates existing testing procedures for predictability. In this paper we propose novel empirical likelihood methods …
EQ-Net combines LLR estimation and quantization using deep learning.
problem Unified solution for LLR estimation and quantization.
method Two-stage algorithm using LLR compression as a pretext task.
result Achieves state-of-the-art results with gains in efficiency and latency.
Sequential hypothesis testing is a desirable decision making strategy in any time sensitive scenario. Compared with fixed sample-size testing, sequential testing is capable of achieving identical probability of error requirements using less samples in average. For a binary detection problem, it is well known that for k…
A new DC programming approach improves RBM training efficiency.
problem Improving the training efficiency of Restricted Boltzmann Machines (RBMs).
method Formulated a stochastic DC programming approach to minimize RBM log-likelihood.
result The new algorithm achieves higher log-likelihood more rapidly with the same computational budget.
This paper develops embeddings that preserve likelihood-based statistical inference.
problem Modern machine learning embeddings destroy the geometric structure required for likelihood-based inference.
method Developed a rigorous theory of likelihood-preserving embeddings and introduced the Likelihood-Ratio Distortion metric.
result Controlling the distortion Δn is necessary and sufficient for preserving inference. We consider the finite sample properties of the regularized high-dimensional Cox regression via lasso. Existing literature focuses on linear models or generalized linear models with Lipschitz loss functions, where the empirical risk functions are the summations of independent and identically distributed (iid) losses. T…
Quantum machine learning uses quantum cross entropy to minimize loss, but measurement loss affects this process.
problem Quantum machine learning's loss minimization through cross entropy is affected by measurement outcomes.
method Defined quantum cross entropy, proved its lower bounds, and investigated its relation to quantum fidelity and likelihood.
result Quantum cross entropy is lower-bounded by negative log-likelihood when derived from quantum data, but measurement outcomes can cause loss.
Invertible DenseNets improve model efficiency and performance.
problem Improving model efficiency and performance in neural networks.
method Enforcing invertibility in DenseNets by satisfying the Lipschitz constraint and proposing a learnable concatenation.
result i-DenseNets outperform Residual Flows in negative log-likelihood on various datasets.
A new hybrid VAE-GAN framework improves mode coverage and quality.
problem Mode collapse and poor sample quality in GANs and VAEs.
method Integrates a 'Best-of-Many-Samples' reconstruction cost and a stable synthetic likelihood estimate.
result Significant improvement in mode coverage and quality compared to hybrid VAE-GANs and plain GANs.
A new method detects changes in multivariate data using random forests.
problem Detecting changes in multivariate data.
method A computationally feasible search method using random forests and class probability predictions.
result Consistently locates change points in simulations.
When the in-sample Sharpe ratio is obtained by optimizing over a k-dimensional parameter space, it is a biased estimator for what can be expected on unseen data (out-of-sample). We derive (1) an unbiased estimator adjusting for both sources of bias: noise fit and estimation error. We then show (2) how to use the adjust…
Partition Tree estimates conditional densities for mixed continuous and categorical variables.
problem Estimating conditional densities for mixed data types.
method Tree-based framework modeling conditional distributions as piecewise-constant densities on adaptive partitions, minimizing conditional negative log-likelihood.
result Improved probabilistic prediction compared to CART-style trees and state-of-the-art methods.
Proposes CLUB for reliable MI minimization in high dimensions.
problem Estimating and minimizing mutual information in high-dimensional spaces.
method Contrastive Log-ratio Upper Bound (CLUB) for MI minimization.
result CLUB provides reliable estimation of mutual information.
ELU algorithm improves on EM for over-specified Gaussian mixtures.
problem Slow convergence of EM in over-specified Gaussian mixtures.
method Developed ELU algorithm for two-component mixtures, combining exponential location update and gradient descent.
result ELU converges to final statistical radius after logarithmic iterations, resolving open question.
DNLL loss improves deep LDA accuracy and consistency.
problem Pathological solutions in unconstrained Deep LDA.
method Introducing Discriminative Negative Log-Likelihood (DNLL) loss.
result Deep LDA trained with DNLL produces clean latent spaces and better calibrated probabilities.
SRFE clarifies KL divergences without unifying learning frameworks.
problem Inductive biases of KL divergences and their limitations.
method Introducing SRFE, a log-moment-based functional of the likelihood ratio.
result SRFE recovers KL divergences as limits and reveals a mean-variance tradeoff.
We consider log-supermodular models on binary variables, which are probabilistic models with negative log-densities which are submodular. These models provide probabilistic interpretations of common combinatorial optimization tasks such as image segmentation. In this paper, we focus primarily on parameter estimation in…
Deep learning compresses L-values for QAM symbols, reducing memory and improving performance.
problem Efficiently storing log-likelihood ratios (L-values) for QAM modulation.
method A deep autoencoder that jointly compresses and reconstructs L-values with a weighted loss function.
result Reduces memory footprint by up to two times with less than 0.1 dB performance loss.
We show that the log-likelihood of several probabilistic graphical models is Lipschitz continuous with respect to the lp-norm of the parameters. We discuss several implications of Lipschitz parametrization. We present an upper bound of the Kullback-Leibler divergence that allows understanding methods that penalize the …
Neural networks approximate likelihood ratios for complex models.
problem Difficulty in computing likelihood ratios for modern models.
method Applying the likelihood ratio trick with neural network classifiers.
result Different neural network setups can approximate likelihood ratios with varying performance.
Direct neural ratio estimator for likelihood-free inference.
problem Efficient likelihood estimation for complex models.
method Amortized likelihood ratio estimation using neural networks.
result DNRE often outperforms previous ratio estimators.
Paper improves likelihood estimation for discrete distributions.
problem Computing profile maximum likelihood for discrete distributions.
method New bounds on Bethe and Sinkhorn permanents for low rank matrices.
result Achieves an approximation factor of exp(-O(sqrt(n) log n)) in polynomial time.
GANs outperform traditional methods in speech source separation.
problem Improving speech source separation accuracy.
method Used a multi-layer perceptron trained with a Wasserstein-GAN formulation to compare against traditional methods.
result Multi-layer perceptron with Wasserstein-GAN outperformed other methods in source to distortion ratio.
New methods use machine learning to constrain particle physics parameters.
problem Constraining effective field theory parameters in collider experiments.
method Inference techniques using latent-space structure and neural networks trained on augmented data.
result Significantly stronger bounds on effective dimension-six operators.
Study on descent algorithms in spiked matrix-tensor models, revealing performance issues and solutions.
problem Analysis of descent algorithms in spiked matrix-tensor models.
method Quantitative analysis using Kac-Rice formula and PDEs for Langevin dynamics.
result Gradient flow dynamics slow down in regions with spurious local minima, while AMP performs well.