Stagewise training outperforms vanilla SGD in accelerating convergence and testing error reduction.
problem Improving the convergence rate of SGD for neural networks.
method Stagewise training strategy with a geometrically decreasing step size, compared to vanilla SGD with a polynomially decaying step size.
result Stagewise training achieves faster convergence and testing error reduction compared to vanilla SGD under the Polyak-Łojasiewicz condition.
Proposes a new method for kernel density estimation using stagewise minimization and a simple dictionary.
problem Kernel density estimation with data-adaptive weighting parameters and sparse representation.
method Stagewise minimization algorithm based on U-divergence and a simple dictionary. result Develops non-asymptotic error bound for the proposed estimator.
This paper addresses convergence issues in non-convex optimization problems using stagewise learning.
problem Theoretical gaps in convergence for non-convex problems and lack of adaptive step size theories.
method Proposes a stagewise optimization framework for non-smooth non-convex problems using adaptive step sizes.
result Demonstrates adaptive convergence of stagewise AdaGrad and improved generalization performance.
Forward stagewise regression follows a very simple strategy for constructing a sequence of sparse regression estimates: it starts with all coefficients equal to zero, and iteratively updates the coefficient (by a small amount ε) of the variable that achieves the maximal absolute inner product with the current residua…
Stagewise boosting improves gradient boosting for distributional regression.
problem Vanishing gradient in gradient boosting for distributional regression leads to suboptimal models.
method Proposes a stagewise boosting-type algorithm for distributional regression, combining stagewise regression ideas with gradient boosting and incorporating a novel regularization method, correlation filtering.
result The proposed algorithm provides better results, especially for complex distributions, by reducing the risk of being trapped in a local optimum.
The performance of EM in learning mixtures of product distributions often depends on the initialization. This can be problematic in crowdsourcing and other applications, e.g. when a small number of 'experts' are diluted by a large number of noisy, unreliable participants. We develop a new EM algorithm that is driven by…
Proposes MM-DUST for efficient generalized lasso solution paths.
problem Efficiently solve generalized lasso problems in large-scale and non-linear models.
method Majorization-minimization dual stagewise algorithm incorporating quadratic majorizers and stagewise learning.
result Established the uniform convergence of approximated solution paths.
Proposes a model to relate a tensor feature to a univariate outcome using sparse and low-rank components.
problem Relating a univariate outcome to a feature tensor with sparse and low-rank components.
method Divide-and-conquer strategy, stagewise estimation procedure for unit-rank tensor regression.
result The stagewise solution paths converge to those of regularized regression as step size goes to zero.
We present a method of variable selection for the sparse generalized additive model. The method doesn't assume any specific functional form, and can select from a large number of candidates. It takes the form of incremental forward stagewise regression. Given no functional form is assumed, we devised an approach termed…
Boosting methods are highly popular and effective supervised learning methods which combine weak learners into a single accurate model with good statistical performance. In this paper, we analyze two well-known boosting methods, AdaBoost and Incremental Forward Stagewise Regression (FSε), by establishing t…
A new method SEBS optimizes SGD batch size for better performance.
problem Optimizing batch size for SGD to balance training speed and generalization.
method SEBS method uses a multi-stage geometric batch size enlargement scheme.
result SEBS reduces parameter updates without increasing generalization error.
A machine learning approach to record fusion with high accuracy.
problem Aggregating multiple records corresponding to the same entity.
method Constructing feature vectors from attribute-level, record-level, and database-level signals; using a stagewise additive model to learn a classifier.
result Average precision of ~98% with source information and ~94% without source information across diverse datasets.
Divide-and-conquer method speeds sparse factorization for large matrices.
problem Sparse factorization of large matrices for statistical learning.
method Statistical problem formulation, divide-and-conquer approach, stagewise learning.
result Efficient algorithm with lower complexity than existing methods.
This paper presents a natural extension of stagewise ranking to the the case of infinitely many items. We introduce the infinite generalized Mallows model (IGM), describe its properties and give procedures to estimate it from data. For estimation of multimodal distributions we introduce the Exponential-Blurring-Mean-Sh…
Least Angle Regression is a promising technique for variable selection applications, offering a nice alternative to stepwise regression. It provides an explanation for the similar behavior of LASSO (ℓ1-penalized regression) and forward stagewise regression, and provides a fast implementation of both. The idea has…
Paper analyzes convergence of memory-based distributed SGD with momentum.
problem High communication cost in distributed SGD for large models.
method Introduces transformation equation to analyze convergence of M-DSGD with momentum.
result Universal convergence analysis for M-DSGD with momentum for convex and non-convex problems.
This paper addresses the general problem of modelling and learning rank data with ties. We propose a probabilistic generative model, that models the process as permutations over partitions. This results in super-exponential combinatorial state space with unknown numbers of partitions and unknown ordering among them. We…
This work provides simple algorithms for multi-class (and multi-label) prediction in settings where both the number of examples n and the data dimension d are relatively large. These robust and parameter free algorithms are essentially iterative least-squares updates and very versatile both in theory and in practice. O…
Estimates density ratio for two-sample comparison using tree models.
problem Comparing two distributions given i.i.d. observations.
method Additive tree models with balancing loss for density ratio estimation.
result Bayesian inference provides uncertainty quantification for density ratio.
Paper tackles robust optimization under uncertainty using nested distance.
problem Optimizing under distributionally robust uncertainty with nested distance.
method Equivalent recursive and dynamic programming reformulations for tractable optimization.
result Optimal robust policies can be found efficiently using convex optimization.
The paper decomposes probabilistic scores into reliability, uncertainty, and information loss.
problem Understanding the reliability and uncertainty of probabilistic predictions.
method Developed decomposition identities for proper losses, quantifying reliability, residual uncertainty, and information gain.
result A three-term identity for classification scores, revealing miscalibration, grouping term, and feature-level uncertainty.
This study improves lookahead Bayesian optimization using rollout approximation.
problem Error propagation in lookahead Bayesian optimization due to model mis-specification.
method Proves rollout's improving nature in lookahead BO and provides guidelines for choosing the rolling horizon.
result Empirical results show rollout improves over myopic and non-myopic BO algorithms.
TSL learns separable models to avoid signal cancellation and off-support extrapolation.
problem Signal cancellation and off-support extrapolation in additive models.
method Tensor Separation Learning (TSL) via stagewise greedy procedure with orthogonal refitting.
result TSL avoids information loss caused by marginalizing higher-order interactions.
CASP selects reliable policies for two-stage recommender systems by considering both value and support.
problem The selection of a generator in two-stage recommender systems affects both the policy value and the data support used to estimate it.
method CASP combines doubly robust value estimation with a support-burden penalty.
result CASP selects lower-burden policies when estimated value and support credibility are in tension.
StageOpt efficiently optimizes safe decisions by separating safety and utility stages.
problem Optimizing unknown utility with safety constraints in sequential decisions.
method Develops StageOpt, a two-stage safe Bayesian optimization algorithm.
result StageOpt is more efficient and applicable to broader problems than existing methods.
Efficiently performs robust and sparse kernel regression.
problem Robust and sparse kernel regression.
method Sign gradient descent and early stopping.
result Sign gradient descent achieves robust and sparse kernel regression efficiently.
REGAIN learns optimal auxiliary directions for forecast reconciliation.
problem Forecast reconciliation from fixed systems; identifying useful auxiliary directions.
method REGAIN learns normalized auxiliary directions, forecasts induced series, and selects directions by loss reduction.
result Gain-selected auxiliary directions improve forecast quality, especially for residual uncertainty.
New method selects variables for GP regression using sparse projection.
problem Identifying environmental factors affecting metal corrosion.
method Sparse projection of input variables, gradient descent optimization, non-convex marginal likelihood.
result Proposed method outperforms benchmarks in variable selection accuracy.
In this paper we analyze boosting algorithms in linear regression from a new perspective: that of modern first-order methods in convex optimization. We show that classic boosting algorithms in linear regression, namely the incremental forward stagewise algorithm (FSε) and least squares boosting (LS-Boost($…
STL-SGD accelerates Local SGD by gradually increasing communication periods.
problem Reducing communication complexity in distributed SGD.
method Gradually increasing communication periods while decreasing learning rate.
result Significant reduction in communication complexity with improved convergence rates.
Algorithm infers sampling distribution from i.i.d. samples without supervision.
problem Learning probability distributions from unlabeled data.
method Unsupervised tree boosting using additive tree ensembles and new distributional operations.
result Algorithm outperforms deep learning in multivariate density estimation.
This paper analyzes l1-regularized PageRank for local graph clustering, proving its effectiveness and efficiency.
problem Local graph clustering in large graphs, focusing on recovering a single target cluster given a seed node.
method Statistical analysis of l1-regularized PageRank method for recovery of a target cluster.
result l1-regularized PageRank recovers the full target cluster with bounded false positives and exactly the target cluster if the seed is connected solely to it.
Taylorized training improves neural network training at finite width.
problem Understanding and improving neural network training at finite width.
method Training the k-th order Taylor expansion of the neural network at initialization.
result Taylorized training agrees with full neural network training better as k increases and can significantly close the performance gap.
Self-training outperforms pre-training on COCO object detection and segmentation datasets.
problem The effectiveness of pre-training in improving object detection and segmentation models is limited.
method Investigated self-training as an alternative method to utilize additional data.
result Self-training consistently improves model performance across various dataset sizes and data augmentation levels.
Un-trained neural networks outperform trained methods in MRI reconstruction.
problem Accelerated MRI reconstruction with minimal training data.
method Variation of Deep Decoder without training data.
result Un-trained approach significantly outperforms other methods in reconstruction accuracy.
Meta-learning performance depends on train-validation split type.
problem Understanding the importance of train-validation split in meta-learning.
method Theoretical and experimental study comparing train-val and train-train methods.
result Train-train method can achieve strictly better excess loss in realizable cases.
New method reveals how training data influence diffusion model outputs.
problem Difficulty in assessing training data impact on diffusion model outputs.
method Use of ensembles trained on carefully engineered splits of training data to identify influential training examples.
result Demonstrated the viability of ensembles as generative models and validity of assessing influence.
Paper shows adversarial training can be fooled by new type of noise.
problem Adversarial training can be fooled by new types of noise.
method Designing ADVIN, a new type of inducing noise.
result ADVIN can degrade adversarial training robustness by 99.9%.
New MIP methods improve training of integer-valued neural networks.
problem Training integer-valued neural networks with limited data and resources.
method Formulated new MIP models to optimize training efficiency and handle more data.
result Significantly outperforms previous state-of-the-art methods in accuracy, training time, and data usage.
Free adversarial training reduces the generalization gap compared to vanilla method.
problem Improving generalization in adversarial training.
method Analysis of algorithmic stability in free adversarial training.
result Free adversarial training shows a lower generalization gap.
ADASS selects adaptive subsets for SGD training acceleration.
problem Fixed sample size in SGD limits training efficiency.
method ADASS selects adaptive subsets based on Lipschitz constants.
result ADASS achieves comparable accuracy with full training set.
Solve-training trains neural nets to map physical solutions efficiently.
problem Representing complex physical solutions with neural networks.
method Variational training using loss functions from physical models.
result Effective neural network representation of solution maps without expensive labels.
MixTrain improves verifiable robustness of neural networks without sacrificing efficiency.
problem Efficiently making neural networks robust against adversarial attacks.
method Stochastic robust approximation and dynamic mixed training techniques.
result Achieves up to 95.2% verified robust accuracy with significantly reduced training time.
learn2mix trains neural nets faster by adjusting class proportions dynamically.
problem Training neural nets efficiently with limited resources and imbalanced classes.
method Adaptive class proportion adjustment during training.
result Neural nets trained with learn2mix converge faster than static methods.
Loss-guided training accelerates node embedding methods on graphs.
problem Training efficiency in graph learning methods with implicit positive examples.
method Dynamic adjustment of training distribution based on loss values.
result Significant acceleration in training and computation over static methods.
Paper explores fast adversarial training to improve robustness with less computation.
problem Efficiently defending against adversarial examples.
method Integrates simple self-attacks for faster training, focusing on overfitting recovery.
result Shows superior robust accuracy with reduced training time compared to strong adversarial training.
Three LF training criteria improve neural network acoustic models without cross-entropy pre-training.
problem Improving purely sequence-trained neural network acoustic models.
method Comparison of three lattice-free discriminative training criteria (MMI, bMMI, sMBR) on LVCSR tasks.
result LF-bMMI models outperform plain LF-MMI models by 5% WER on Switchboard datasets.
Free adversarial training improves robustness without generating adversarial examples.
problem Training robust models against adversarial attacks is costly and impractical for large-scale datasets.
method Recycles gradient information from parameter updates to generate adversarial examples.
result Free adversarial training achieves comparable robustness to PGD training at negligible cost.