Partial model averaging improves Federated Learning performance.
problem Periodic model averaging causes significant model discrepancy in Federated Learning.
method Proposes a partial model averaging framework that encourages local models to stay close to each other.
result Partial averaging achieves up to 2.2% higher validation accuracy than full averaging.
Proposes a new framework for balancing average- and worst-case performance in machine learning.
problem Robustness issues in machine learning, especially in safety-critical domains.
method Probabilistic robustness framework that balances average- and worst-case performance.
result Effective algorithm balances average- and worst-case performance with lower computational cost.
WASH trains ensembles with shuffled weights to improve accuracy and reduce communication.
problem Training ensembles for weight averaging leads to models converging to different loss basins.
method WASH randomly shuffles a small percentage of weights during training to keep models within the same basin.
result WASH achieves state-of-the-art image classification accuracy with lower communication costs.
Nested model averaging improves high-dimensional linear regression performance.
problem High-dimensional linear regression with predictor ordering impact.
method Combining model averaging with regularized estimators on the solution path.
result Nested model averaging with lasso and SLOPE outperforms competing methods.
Tail averaging consists in averaging the last examples in a stream. Common techniques either have a memory requirement which grows with the number of samples to average, are not available at every timestep or do not accomodate growing windows. We propose two techniques with a low constant memory cost that perform tail …
The study examines how averaging data improves model performance.
problem Understanding the generalization gap in machine learning models.
method Data averaging, covariance analysis, and stochastic gradient descent (SGD) noise modeling.
result A modified generalization gap is always non-negative for a large class of model parameter distributions.
Paper introduces stability in model averaging and proposes a L2-penalty method.
problem Theoretical properties of model averaging from stability perspective.
method Introduces stability, defines asymptotic empirical risk minimizer, and proposes L2-penalty model averaging method.
result Proposed L2-penalty method ensures stability and consistency under reasonable conditions.
This paper studies the correlations of the average winnings of agents and the volatilities of systems based on mix-game model which is an extension of minority game (MG). In mix-game, there are two groups of agents; group1 plays the majority game, but the group2 plays the minority game. The results show that the correl…
Study proposes worst+gap measure for better DG evaluation.
problem Lack of comprehensive exploration of average measure in DG evaluation.
method Introduced worst+gap measure and compared it with average measure.
result Worst+gap measure provides a more accurate approximation of true DG performance.
The paper tackles batch policy learning in Markov Decision Processes, focusing on average reward maximization.
problem Maximizing long-term average reward in Markov Decision Processes with batch learning.
method Doubly robust estimator for average reward, optimization algorithm for optimal policy, finite-sample regret guarantee.
result The proposed method achieves semiparametric efficiency and provides a finite-sample regret guarantee.
PACE optimizes training for averaged language models, improving performance.
problem How to optimize training for averaged language model iterates.
method Formulated as an optimal-control problem, solved for minimizing error of the average with a penalty on intervention size.
result PACE improves the limiting squared error of the iterate-average estimator by an arbitrarily large factor on some instances.
Most systems and learning algorithms optimize average performance or average loss -- one reason being computational complexity. However, many objectives of practical interest are more complex than simply average loss. This arises, for example, when balancing performance or loss with fairness across people. We prove tha…
Enhances Fourier estimator performance for asynchronous event-data.
problem Improving correlation and covariance estimation on event-data.
method Implement and test NUFFT methods with different averaging kernels.
result Demonstrates improved performance and relationship between averaging scales.
Averaged SGD optimizes a smoothed objective, leading to better generalization.
problem Improving generalization performance in machine learning models.
method Analyzed the smoothed objective function of SGD and proved that averaged SGD can optimize this smoothed function efficiently.
result Averaged SGD can efficiently optimize a smoothed objective, leading to better generalization.
Instability and variability of Deep Reinforcement Learning (DRL) algorithms tend to adversely affect their performance. Averaged-DQN is a simple extension to the DQN algorithm, based on averaging previously learned Q-values estimates, which leads to a more stable training procedure and improved performance by reducing …
New RL algorithms improve average-reward performance.
problem Improving RL algorithms for average-reward criteria.
method Developed novel algorithms addressing average-reward criterion directly.
result ATRPO significantly outperforms TRPO in challenging MuJuCo environments.
New algorithms optimize spectral risk measures, improving interpolation between average and worst-case performance.
problem Optimizing spectral risk measures for learning systems.
method Developed stochastic algorithms to optimize spectral risk measures by characterizing their subdifferential and addressing challenges like biasedness of subgradient estimates and non-smoothness.
result Our approach outperforms out-of-the-box stochastic subgradient and dual averaging methods in optimizing spectral risk measures.
This work proposes a model averaging method for SVM that avoids redundant covariates and achieves asymptotic optimality.
problem Redundant covariates impair SVM performance in high-dimensional settings.
method Frequentist model averaging procedure for SVM using cross-validation to select optimal weights.
result The proposed method achieves asymptotic optimality in SVM model averaging.
Bayesian stacking improves model performance with varying model weights.
problem Improving model predictions with heterogeneous input performance.
method Bayesian hierarchical stacking with varying model weights inferred via Bayesian inference.
result Hierarchical stacking yields better predictions than linear averaging.
SQR Averaging improves probabilistic electricity price forecasting.
problem Accurate short-term price forecasting in electricity markets.
method Smoothing Quantile Regression Averaging.
result SQR Averaging leads to profit increases of up to 3.5% in day-ahead power trading.
Study improves reinforcement learning for stable long-term performance.
problem Distributionally robust average-reward reinforcement learning for stable long-term performance.
method Proposes two algorithms to achieve near-optimal sample complexity.
result Achieves a sample complexity of O(∣S∣∣A∣tmix2ε−2) for estimating optimal policy and robust average reward. Study ranks of elliptic curves via prime averages.
problem Classifying elliptic curves by rank.
method Average Frobenius trace over primes, data science experiments.
result Oscillating pattern in average trace values, correlates with rank.
High-dimensional models can outperform simpler ones in causal inference.
problem Estimating average treatment effects with many covariates.
method High-dimensional linear regression and synthetic control with many control units.
result Adding more control units can improve imputation performance even when pre-treatment fit is perfect.
In high dimension, low sample size (HDLSS)settings, the simple average distance classifier based on the Euclidean distance performs poorly if differences between the locations get masked by the scale differences. To rectify this issue, modifications to the average distance classifier was proposed by Chan and Hall (2009…
A new framework evaluates deep learning vs classical forecasting methods for time series predictions.
problem Current forecasting model evaluation metrics fail to capture model performance differences.
method Proposes a novel framework for evaluating univariate time series forecasting models from multiple perspectives.
result Deep learning models like NHITS outperform classical methods in multi-step ahead forecasting but not in anomaly handling.
KL regularization helps RL algorithms by implicitly averaging q-values.
problem Understanding why KL regularization improves RL performance.
method An approximate value iteration scheme, studying KL and entropy regularization.
result Strong performance bound combining linear horizon dependency and averaging effect of estimation errors.
Bayesian model averaging has become a widely used approach to accounting for uncertainty about the structural form of the model generating the data. When data arrive sequentially and the generating model can change over time, Dynamic Model Averaging (DMA) extends model averaging to deal with this situation. Often in ma…
Paper develops bounds for stochastic approximation with averaging.
problem Establish high-probability bounds for averaged stochastic approximation.
method Develops a general framework for non-asymptotic concentration bounds.
result Derives sharp bounds for averaged iterates and tightens existing results.
Distributed statistical learning problems arise commonly when dealing with large datasets. In this setup, datasets are partitioned over machines, which compute locally, and communicate short messages. Communication is often the bottleneck. In this paper, we study one-step and iterative weighted parameter averaging in s…
This paper introduces sample-averaged Q-learning for better RL performance.
problem Improving reinforcement learning algorithms by managing uncertainty.
method Integrates statistical inference into Q-learning through sample averaging and functional central limit theorem.
result Establishes a unified theoretical foundation for sample-averaged Q-learning.
Novel approach simplifies VI problems with faster performance.
problem Black-box VI optimization problems.
method Sample Average Approximation (SAA) combined with quasi-Newton methods and line search.
result Achieves faster performance than existing methods.
Paper predicts cryptocurrency bull and bear phases using Bitcoin's moving averages.
problem Determining cryptocurrency bull and bear phases based on Bitcoin performance.
method Employing predictive algorithms to forecast Bitcoin's 50 Day and 200 Day Moving Averages.
result Predicted data from Bitcoin's moving averages helps identify potential bull and bear phases.
Bayesian model averaging fails under covariate shift, affecting neural networks' performance.
problem Bayesian model averaging's failure in neural networks under covariate shift.
method Explained the issue and proposed novel priors to improve robustness.
result Bayesian model averaging is problematic under covariate shift, especially with linear feature dependencies.
A new method for averaging model predictions using minimum divergence.
problem Improving model averaging methods, especially in small samples.
method Minimum divergence framework for model weight calculation.
result Empirically outperforms standard model averaging methods.
Deterministic method for certifying neural network robustness.
problem Certifying neural network robustness against adversarial attacks.
method Equivalence between training and Gaussian averaging for robustness certification.
result Comparable certified accuracy and robustness to stochastic methods but with single model evaluation.
This paper improves forecasts for diverse time series by averaging similar ones.
problem Forecasting challenges in heterogeneous time series.
method Dynamic Time Warping to find similar time series, k-Nearest Neighbor averaging.
result Averaging improves forecasts of simple models.
There are non-vanishing price responses across different stocks in correlated financial markets. We further study this issue by performing different averages, which identify active and passive cross-responses. The two average cross-responses show different characteristic dependences on the time lag. The passive cross-r…
Improved generalization with iterate averaging and adaptive algorithms.
problem Enhancing model generalization performance in high-dimensional settings.
method Gaussian process perturbation model, combine IA with large learning rates and regularisation, less frequent averaging, adaptive gradient methods.
result Iterate averaging with adaptive algorithms significantly outperforms SGD on various datasets.
Revises Bayesian model averaging for foundation models.
problem Ensemble pre-trained and lightly-finetuned foundation models for improved classification performance.
method Introduces trainable linear classifiers and computationally cheaper model averaging scheme (OMA).
result Ensembled models can better predict on various datasets.
In this paper, we focus on quantifying model stability as a function of random seed by investigating the effects of the induced randomness on model performance and the robustness of the model in general. We specifically perform a controlled study on the effect of random seeds on the behaviour of attention, gradient-bas…
Two-Tailed Averaging improves generalization by optimizing the number of leading iterates to ignore.
problem Improving generalization in stochastic optimization with limited resources and hyperparameters.
method An anytime adaptive algorithm that balances the number of leading iterates to ignore for better generalization.
result Approximates the optimal tail at all optimization steps, improving generalization without hyperparameters.
This work explores alternative learning criteria beyond traditional risk.
problem The trade-offs of optimizing for average performance in machine learning.
method Survey and introduction of non-traditional learning criteria.
result Emphasizes the importance of considering the loss distribution for desirable performance.
Improving optimization for iterate-averaged language models
problem How to optimize the averaged model returned by Language Model pipelines
method Formulating optimizer design as an optimal-control problem
result Proven convergence rate and strict improvement in squared error
Two new ensemble methods improve CATE estimation across various scenarios.
problem Estimating CATE in clinical trials to understand treatment effects heterogeneity.
method Proposed two ensemble methods: Stacked X-Learner and Consensus Based Averaging (CBA).
result Ensemble methods achieve good performance across diverse scenarios.
New approach shapes error distribution in long-term forecasting.
problem Disparate error distributions in recent transformer models.
method Loss shaping constraints to respect upper bounds on loss at each time-step.
result Competitive average performance with shaped error distribution.
Ensemble learning is a method of combining multiple trained models to improve model accuracy. We propose the usage of such methods, specifically ensemble average, inside Convolutional Neural Network (CNN) architectures by replacing the single convolutional layers with Inner Average Ensembles (IEA) of multiple convoluti…
New method SF-AdamW trains large models without decay phases or memory overhead.
problem Inadequate fixed compute budgets for large-scale training.
method Schedule-Free (SF) method revisited and refined.
result SF-AdamW effectively navigates loss landscape without decay phases or memory overhead.
MDA optimizer performs similarly to SGD+M in CV and Adam in NLP.
problem Performance degradation due to choosing the wrong optimizer.
method Modernized Dual Averaging (MDA) optimizer, inspired by dual averaging.
result MDA performs as well as SGD+M in CV and as Adam in NLP.