Empirical law predicts accuracy of Google Translate's translation chains.
problem Predicting accuracy in machine translation with multiple hops.
method Empirical testing of Google Translate's sequential translation.
result Accuracy decreases with the number of translating hops, following a power law.
We identify and optimize the fairness-accuracy tradeoff through TAF Curves and FAUC metrics.
problem Balancing fairness and accuracy in machine learning models for high-stakes decisions.
method Developed TAF Curves and FAUC metric to quantify the tradeoff, and introduced FairStacks framework to expand the Pareto frontier.
result FairStacks framework expands the empirical Pareto frontier and improves the FAUC for model ensembles.
Neural network accuracy improves with denser training samples.
problem Improving neural network accuracy on unseen test samples.
method Bounding empirical training error smoothed across activation regions and using it to discard high-risk test samples.
result Discarding high-risk test samples based on error bounds improves prediction accuracy by up to 20%.
EMPIR combines low and full precision DNNs to enhance robustness against adversarial attacks.
problem Vulnerability of DNNs to adversarial attacks that misclassify inputs with small perturbations.
method Ensemble of quantized DNN models with different numerical precisions.
result EMPIR ensembles increase adversarial robustness by 42.6% on average across different tasks.
Bayesian predictive inference analyzes a dataset to make predictions about new observations. When a model does not match the data, predictive accuracy suffers. We develop population empirical Bayes (POP-EB), a hierarchical framework that explicitly models the empirical population distribution as part of Bayesian analys…
Study shows accuracy of neural networks depends more on error location than percentage of error.
problem Effect of noise on accuracy in incremental learning neural networks.
method Empirical study using Perceptron, Feed Forward Neural Network, and Radial Basis Function Neural Network.
result Accuracy of neural networks is more dependent on error location than the percentage of error.
JTT improves model worst-group accuracy without group annotations.
problem Low worst-group accuracy in standard ERM models with spurious correlations.
method Two-stage approach: first ERM, then upweight misclassified examples.
result JTT closes 75% of the gap in worst-group accuracy compared to group DRO.
A new method for selective classification trades off accuracy for coverage.
problem Selective classification allows a classifier to abstain from predicting some instances.
method Optimizes a collection of class-wise decoupled one-sided empirical risks.
result The method achieves near-optimal coverage in high target accuracy regimes.
Mixup improves model accuracy and calibration through data transformation and random perturbation.
problem Improving model accuracy and calibration in machine learning.
method Interprets Mixup as empirical risk minimization with data transformation and random perturbation.
result Mixup induces multiple known regularization schemes that prevent overfitting and overconfident predictions.
Deep neural networks have achieved impressive performance in many applications but their large number of parameters lead to significant computational and storage overheads. Several recent works attempt to mitigate these overheads by designing compact networks using pruning of connections. However, we observe that most …
Adversarial robustness has become an important research topic given empirical demonstrations on the lack of robustness of deep neural networks. Unfortunately, recent theoretical results suggest that adversarial training induces a strict tradeoff between classification accuracy and adversarial robustness. In this paper,…
Compensation methods correct overestimation of adversarial robustness in neural networks.
problem Overestimation of adversarial robustness using first-order attack methods.
method Proposed compensation methods address inaccurate gradient computation and reduce backpropagations.
result Empirical evaluation of adversarial robustness is improved with these methods.
The paper analyzes adversarial training effects on classification accuracy.
problem Understanding adversarial training's impact on standard and robust accuracy.
method Derived precise statistical analysis for binary classification problems with Gaussian data.
result Theoretical explanation of standard and robust accuracy trends for adversarial training.
The paper examines how adversarial robustness affects accuracy disparity across different classes.
problem Understanding the impact of adversarial robustness on accuracy disparity across different classes.
method Linear classifiers under a Gaussian mixture model, decomposing the impact into inherent and imbalance effects.
result Adversarial robustness consistently degrades standard accuracy in balanced classes, but the class imbalance ratio plays a different role in accuracy disparity.
Extracurricular learning closes the accuracy gap in knowledge distillation.
problem Accuracy gap between teacher and student models after knowledge distillation.
method Modeling student and teacher output distributions, sampling from an extended data distribution, and matching over this set.
result Extracurricular learning reduces the accuracy gap by 46% to 68%.
Unified smoothing for robust classification improves accuracy.
problem Improving robustness of classifiers against adversarial attacks.
method Learned smoothed densities and randomized smoothing.
result Provable robust accuracies higher than state-of-the-art defenses.
SCORE resolves the robustness vs accuracy trade-off by redefining robust error.
problem The inherent trade-off between robustness and accuracy in adversarial training.
method SCORE defines local equivariance as the ideal robust behavior, leading to a new robust error metric.
result SCORE reconciles robustness and accuracy, improving model performance on RobustBench.
Improved accuracy with fewer labels using MixMatch and active learning.
problem Achieving high accuracy with limited labeled data.
method Combining MixMatch and active learning techniques.
result Significant improvement in accuracy on CIFAR-10, CIFAR-100, and SVHN datasets.
Novel hybrid method for Bayesian network structure learning reduces computational time without sacrificing accuracy.
problem Bayesian network structure learning efficiency and accuracy trade-off.
method Partitioned PC (pPC), p-value adjacency thresholding (PATH), hybrid greedy initialization (HGI). result pHGS achieves significant computational reductions compared to the PC algorithm without sacrificing structure learning accuracy.
Auto-decoder synthesizes graphs from latent codes.
problem Creating new graph structures from specified distributions.
method Generative model learns latent codes from empirical distribution. Self-attention identifies likely connectivity patterns. Graph-based normalizing flows sample latent codes.
result Model outperforms state of the art by 1.5x in accuracy and 2x in speed.
This paper improves forest pruning to balance accuracy and interpretability.
problem Limited interpretability of regression forests.
method Lasso-pruning and theoretical analysis of regression forests.
result Pruned regression forests can achieve equal or better accuracy than unpruned ones, with significant size reduction.
ROI-driven data analytics guides investment in empirical data analysis.
problem Determining the optimal depth and breadth of data analytics.
method Conceptual framework validated through empirical studies focusing on dependency extraction in Mozilla Firefox project.
result ROI-driven data analytics helps avoid over-analyzing empirical data.
We demonstrate, theoretically and empirically, that adversarial robustness can significantly benefit from semisupervised learning. Theoretically, we revisit the simple Gaussian model of Schmidt et al. that shows a sample complexity gap between standard and robust classification. We prove that unlabeled data bridges thi…
In this paper we focus on the problem of assigning uncertainties to single-point predictions. We introduce a cost function that encodes the trade-off between accuracy and reliability in probabilistic forecast. We derive analytic formula for the case of forecasts of continuous scalar variables expressed in terms of Gaus…
A new hierarchical forecasting method improves overall accuracy.
problem Hierarchical forecasting challenges, especially for intermittent time series.
method Top-down alignment of independent level forecasts using deep learning and tree-based algorithms.
result Improves overall forecasting accuracy compared to existing methods.
Fairness constraints can improve accuracy from biased data.
problem Learning from biased training data can produce biased and suboptimal classifiers.
method Examined fairness-constrained ERM and other recovery methods.
result Equal Opportunity fairness constraint combined with ERM provably recovers Bayes Optimal Classifier under various bias models.
Proposes a private empirical bootstrap for Gaussian Differential Privacy.
problem Quantifying uncertainty in massive data under Differential Privacy.
method Gaussian Differential Private Bootstrap by Subsampling.
result Consistent and efficient private inference method.
New analysis shows transformer models can't learn effectively.
problem Transformer models learning in context can't achieve general predictive accuracy.
method Empirical evidence and mathematical analysis of transformer architecture limitations.
result Transformers cannot achieve general predictive accuracy due to architectural limitations.
New latent variable model improves inflation forecasting accuracy.
problem Improving medium-term inflation forecasting accuracy.
method Formulated and tested a latent variable Phillips curve hypothesis using 3,968 factor combinations.
result Latent variable PC models outperform traditional models by 6-8 quarters.
Linear trends in classifier accuracy observed under distribution shift.
problem Understanding why classifier accuracies show linear trends under distribution shift.
method Assumed model similarity and verified empirically.
result Linear trend in classifier accuracy occurs unless distribution shift is large.
Proves accuracy guarantees for self-supervised learning with correlated positive pairs.
problem Lack of theoretical guarantees for self-supervised learning with correlated positive pairs.
method Novel augmentation graph concept and spectral decomposition loss.
result Provably accurate features under linear probe evaluation.
Deep learning (DL) creates impactful advances following a virtuous recipe: model architecture search, creating large training data sets, and scaling computation. It is widely believed that growing training sets and models should improve accuracy and result in better products. As DL application domains grow, we would li…
Transfer learning have been frequently used to improve deep neural network training through incorporating weights of pre-trained networks as the starting-point of optimization for regularization. While deep transfer learning can usually boost the performance with better accuracy and faster convergence, transferring wei…
The paper shows FtU can reduce discrimination without sacrificing accuracy.
problem Discrimination in machine learning predictions.
method Theoretical and empirical analysis of FtU, connecting with Model Multiplicity.
result FtU can reduce discrimination without reducing accuracy.
MEVA aggregates model predictions to improve accuracy without needing model details.
problem Improving model accuracy by combining multiple models.
method Non-intrusive, data-driven framework that treats models as black boxes and optimizes aggregation methods.
result MVA outperforms MEA in estimating aggregated predictions, enhancing robustness and accuracy.
We study the Nonparametric Maximum Likelihood Estimator (NPMLE) for estimating Gaussian location mixture densities in d-dimensions from independent observations. Unlike usual likelihood-based methods for fitting mixtures, NPMLEs are based on convex optimization. We prove finite sample results on the Hellinger accurac…
The best-known and most commonly used distribution-property estimation technique uses a plug-in estimator, with empirical frequency replacing the underlying distribution. We present novel linear-time-computable estimators that significantly "amplify" the effective amount of data available. For a large variety of distri…
Fair classification has been a topic of intense study in machine learning, and several algorithms have been proposed towards this important task. However, in a recent study, Friedler et al. observed that fair classification algorithms may not be stable with respect to variations in the training dataset -- a crucial con…
Proposes a new method for deep ensembles that improves accuracy and calibration.
problem Improving accuracy and calibration of deep ensembles.
method Estimates confusion matrices of ensemble members and weighs them according to their inferred performance.
result Empirically shows superiority of soft Dawid Skene over ensemble averaging.
The increasing amount of data in astronomy provides great challenges for machine learning research. Previously, supervised learning methods achieved satisfactory recognition accuracy for the star-galaxy classification task, based on manually labeled data set. In this work, we propose a novel unsupervised approach for t…
Study examines how imputation accuracy affects prediction accuracy in regression problems with missing covariates.
problem Missing covariates in regression or classification problems.
method Simulation and empirical analysis using UCI datasets and statistical inference.
result Imputation accuracy impacts prediction accuracy, especially with Machine Learning methods.
A new method sorts models to find the best one with minimal risk.
problem Finding the best model with minimal risk among nested models.
method Nested Empirical Risk (NER) and Sorted NER (S-NER) methods.
result The S-NER method selects the true model order and the most parsimonious model.
We empirically investigate the (negative) expected accuracy as an alternative loss function to cross entropy (negative log likelihood) for classification tasks. Coupled with softmax activation, it has small derivatives over most of its domain, and is therefore hard to optimize. A modified, leaky version is evaluated on…
Improved ADMM for convex distributed learning with differential privacy.
problem Privacy concerns in distributed learning with sensitive data.
method Approximate multi-step ADMM with calibrated noise.
result Higher utility and error bounds asymptotic to state-of-the-art.
Study on trade-offs between accuracy and interpretability in machine learning.
problem Lack of formal study on statistical cost of interpretability.
method Modeling interpretability as a constraint in empirical risk minimization for binary classification.
result Explains conditions under which accuracy trade-off occurs with interpretability constraints.
This research analyzes deep PDE solvers for option pricing accuracy.
problem Understanding the accuracy of deep learning methods for solving PDEs in option pricing.
method Comparative experiments with two neural network algorithms in Black--Scholes and Heston models.
result Empirical convergence rates and training times of TDGF method determined.
Study investigates one-shot semi-supervised learning for image classification.
problem Training deep networks requires many labeled samples, limiting adoption.
method Empirical investigation of FixMatch method for one-shot semi-supervised learning.
result Uneven class accuracy is a barrier to high performance in one-shot semi-supervised learning.
Boosts test performance with importance sampling for subpopulation issues.
problem Limited ERM performance on data with spurious correlation or subpopulation.
method Importance sampling as a tool to solve subpopulation problems.
result Achieves state-of-the-art performance on benchmark datasets.