New causal models perform poorly when evaluated on biased training sets.
problem Sample selection bias affects the evaluation of causal models' prediction performance.
method Re-evaluated prediction performance of causal models on a genetic perturbation data set, proposing a less-biased evaluation set.
result Causal models have similar or worse performance when evaluated on a less-biased set compared to standard association-based estimators.
New approach tackles decision-making under predictions that shape outcomes.
problem Challenges in learning optimal decision rules when predictions influence outcomes.
method Introduces performative omniprediction, a predictor that encodes optimal decision rules for multiple objectives.
result Efficient performative omnipredictors exist under a natural restriction of outcome performativity.
This paper compares preprocessing techniques for XGBoost models on various data sets.
problem Improving predictive performance of XGBoost models through optimal data preprocessing.
method Comparison of feature selection, categorical handling, and null imputation methods.
result XGBoost importance by gain is the most consistent and highest-performing method for feature selection.
Proposes a new cross-validation method to estimate model performance.
problem The standard cross-validation method does not accurately estimate the performance of the recommended model.
method Develops a new random-effects model framework to improve naive cross-validation estimators.
result Proposed estimators outperform conventional and naive methods in estimating model performance.
Improves test set performance and reduces out-of-sample disappointment for unstable models.
problem Ensuring strong test set performance via cross-validation for unstable models.
method Nested k-fold cross-validation with hyperparameter selection based on a weighted sum of cross-validation metric and model stability measure.
result Improves out-of-sample MSE for sparse ridge regression and CART by 4% and 2% respectively, compared to k-fold cross-validation.
We train a network to generate mappings between training sets and classification policies (a 'classifier generator') by conditioning on the entire training set via an attentional mechanism. The network is directly optimized for test set performance on an training set of related tasks, which is then transferred to unsee…
New method improves NN performance across various settings.
problem Improving neural network performance across different datasets and architectures.
method Population Gradients (PG) method to calculate non-local gradient estimates.
result Significantly improves final performance across architectures, data-sets, and hyper-parameters.
IVF k-means algorithm improves performance on large sparse data sets.
problem Efficiently clustering large-scale sparse data sets with numerous classes.
method Sparse data representation and inverted-file structure for high-speed and low-memory clustering.
result IVF achieves better performance than other algorithms on real document data sets.
Deep learning improves image reconstruction, but scaling up training sets doesn't significantly boost performance.
problem Understanding the impact of training set size on deep learning image reconstruction.
method Empirical study and analytical characterization of performance scaling laws.
result Scaling up training set size does not significantly improve reconstruction quality for deep learning.
Study on federated learning with private label sets, showing privacy benefits without significant accuracy loss.
problem Effects of label set heterogeneity and privacy constraints in federated learning.
method Apply classical classifier combination methods and adapt FL methods for private label sets, compare public and private settings.
result Reducing labels harms model performance, but centralized tuning can help.
In this paper, we consider asymptotic properties of the support vector machine (SVM) in high-dimension, low-sample-size (HDLSS) settings. We show that the hard-margin linear SVM holds a consistency property in which misclassification rates tend to zero as the dimension goes to infinity under certain severe conditions. …
We propose improving the privacy properties of a dataset by publishing only a strategically chosen "core-set" of the data containing a subset of the instances. The core-set allows strong performance on primary tasks, but forces poor performance on unwanted tasks. We give methods for both linear models and neural networ…
One significant challenge to scaling entity resolution algorithms to massive datasets is understanding how performance changes after moving beyond the realm of small, manually labeled reference datasets. Unlike traditional machine learning tasks, when an entity resolution algorithm performs well on small hold-out datas…
State-of-the-art machine learning frameworks support a wide variety of design features to enable a flexible machine learning programming interface and to ease the programmability burden on machine learning developers. Identifying and using a performance-optimal setting in feature-rich frameworks, however, involves a no…
Empirical median performs well in estimating location with varying scales.
problem Estimating location with varying scales in data.
method Analysis of empirical median as an estimator.
result Matching upper and lower bounds on estimation error.
This paper presents a study on discriminative artificial neural network classifiers in the context of open-set speaker identification. Both 2-class and multi-class architectures are tested against the conventional Gaussian mixture model based classifier on enrolled speaker sets of different sizes. The performance evalu…
Random small feature subsets outperform FS in diverse datasets.
problem The significance of selected features in high-dimensional datasets is questionable.
method Analysis of 28 diverse datasets (microarray, RNA-Seq, etc.).
result Any arbitrary set of features performs as well as or better than selected features across datasets.
We consider the task of estimating a Gaussian graphical model in the high-dimensional setting. The graphical lasso, which involves maximizing the Gaussian log likelihood subject to an l1 penalty, is a well-studied approach for this task. We begin by introducing a surprising connection between the graphical lasso and hi…
Improved R-QDA classifier performs well in unbalanced data settings.
problem High sensitivity of R-QDA to covariance matrix estimation noise in unbalanced data.
method Proposes an improved R-QDA with two regularization parameters and a modified bias.
result Significantly better classification performance compared to traditional R-QDA.
Algorithms often have tunable parameters that impact performance metrics such as runtime and solution quality. For many algorithms used in practice, no parameter settings admit meaningful worst-case bounds, so the parameters are made available for the user to tune. Alternatively, parameters may be tuned implicitly with…
This work introduces a method to attribute model performance drops to distribution shifts.
problem Attributing performance drops of machine learning models to distribution shifts.
method Formulated as a cooperative game, value of a set of distributions is defined as the change in model performance when only that set of distributions changes. Importance weighting method for computing the value of an arbitrary set of distributions is derived. Quantifying the contribution of each distribution as its Shapley value.
result Demonstrated the effectiveness of the method on various case studies.
New method detects when models influence their own drift in real-time data streams.
problem Models can induce concept drift in real-time data streams.
method CheckerBoard Performative Drift Detection (CB-PDD)
result CB-PDD effectively detects performative drift in real-time data streams.
New algorithm for ML models in gradually adapting data settings.
problem Training models when data distribution reacts to the model over time.
method Stateful Performative Gradient Descent (Stateful PerfGD)
result Stateful PerfGD minimizes performative loss in gradually adapting data settings.
Estimates neural architecture performance speedily.
problem Accurately evaluating neural architectures' generalization performance.
method Estimates final test performance based on training speed.
result Consistently outperforms other alternatives in correlation with true test performance.
Paper studies robust MDPs, improving sample complexity and asymptotic performance.
problem Optimal robust policy and value function in robust MDPs with generative models.
method Improves prior results on non-asymptotic and asymptotic performances of robust MDPs, considering various uncertainty sets.
result Improved sample complexity and asymptotic normality of optimal robust value function.
Study evaluates 31 performance predictors in NAS, recommending best for different settings.
problem Understanding and comparing different performance prediction techniques in NAS.
method Analysis of 31 techniques, testing correlation and rank-based measures, speed-up potential.
result Certain predictor families can be combined for better predictive power.
This work bridges continual learning, active learning, and open set recognition in deep neural networks.
problem Protecting previously acquired representations from catastrophic forgetting in deep neural networks.
method Surveying the literature and proposing a consolidated view to integrate open set recognition and active learning principles.
result Joint improvement in alleviating catastrophic forgetting, querying data, selecting task orders, and robust open world application.
Proposes PUUPL for PUL in imbalanced datasets, boosting minority class signals.
problem Imbalanced datasets and model calibration in PUL.
method Uncertainty-aware pseudo-labeling procedure (PUUPL).
result Substantial performance gains in highly imbalanced settings.
We have developed a novel prediction method based on string invariants. The method does not require learning but a small set of parameters must be set to achieve optimal performance. We have implemented an evolutionary algorithm for the parametric optimization. We have tested the performance of the method on artificial…
Background: Choosing the most performing method in terms of outcome prediction or variables selection is a recurring problem in prognosis studies, leading to many publications on methods comparison. But some aspects have received little attention. First, most comparison studies treat prediction performance and variable…
We investigated the impact of noisy linguistic features on the performance of a Japanese speech synthesis system based on neural network that uses WaveNet vocoder. We compared an ideal system that uses manually corrected linguistic features including phoneme and prosodic information in training and test sets against a …
New meta-score EPP interprets model performance differences.
problem Lack of interpretable benchmarks for model performance.
method Elo-based Predictive Power (EPP) meta-score, logistic regression.
result EPP scores have probabilistic interpretation and can be compared between data sets.
Estimates policy performance in small-data settings without sacrificing data.
problem Poor performance of cross-validation in small-data optimization.
method Uses sensitivity analysis to estimate gradient of optimal objective value.
result Explicit high-probability bounds on error of estimator for small-data, large-scale problems.
StackGenVis simplifies ensemble learning by visualizing model selection and performance.
problem Complexity in choosing and combining models for stacking ensemble learning.
method Visual analytics system that dynamically adapts performance metrics, manages data instances, selects algorithms, and measures predictive performance.
result Reduces complexity of stacking by removing overpromising and underperforming models.
SRN improves set representations for relational reasoning.
problem Set permutational invariance limitations in existing approaches.
method Proposed a Set Refiner Network (SRN) to respect set invariance.
result Substantial gains in prediction performance and robustness on relational reasoning tasks.
This paper considers the challenge of evaluating a set of classifiers, as done in shared task evaluations like the KDD Cup or NIST TREC, without expert labels. While expert labels provide the traditional cornerstone for evaluating statistical learners, limited or expensive access to experts represents a practical bottl…
Feature selection aims to select the smallest subset of features for a specified level of performance. The optimal achievable classification performance on a feature subset is summarized by its Receiver Operating Curve (ROC). When infinite data is available, the Neyman- Pearson (NP) design procedure provides the most e…
A new method reduces hyperparameter tuning evaluations by using sequential tests.
problem Time-consuming hyperparameter tuning in machine learning.
method Sequential Random Search (SQRS) extending regular random search.
result SQRS finds similarly well-performing parameter settings with fewer evaluations.
AUC is unreliable in rare event settings but stable with moderate numbers of events.
problem Misleading performance metrics in rare event settings.
method Simulation study varying dataset sizes and event rates.
result AUC is unreliable in rare event settings but stable with moderate numbers of events.
Data splitting enhances model performance in overparametrized ridgeless regression.
problem Computational inefficiency in training models with large datasets.
method Data splitting as a regularization technique in overparametrized ridgeless regression.
result Data splitting improves statistical performance and computational complexity.
Consider a general machine learning setting where the output is a set of labels or sequences. This output set is unordered and its size varies with the input. Whereas multi-label classification methods seem a natural first resort, they are not readily applicable to set-valued outputs because of the growth rate of the o…
MetaStackVis aids in choosing better metamodels for stacking ensembles.
problem Difficulty in selecting optimal metamodels for stacking ensembles.
method Interactive visualization tool to explore and compare different metamodels.
result Alternative metamodels significantly improve stacking ensemble performance.
Learn2Evaluate uses learning curves to estimate high-dimensional prediction performance.
problem Estimating test performance in high-dimensional data settings is challenging.
method Learn2Evaluate uses learning curves to estimate test performance at the total sample size.
result Learn2Evaluate provides a lower confidence bound for performance estimation.
Automatically adjusts model size for continual Gaussian processes.
problem Determining optimal model size for incremental data.
method Automatic adjustment of model size during continual learning.
result Near-optimal performance across various datasets without hyperparameter tuning.
Early stopping is a widely used technique to prevent poor generalization performance when training an over-expressive model by means of gradient-based optimization. To find a good point to halt the optimizer, a common practice is to split the dataset into a training and a smaller validation set to obtain an ongoing est…
Guarantees for third-person imitation learning from offline data.
problem Improving generalizability in imitation learning.
method Problem-dependent statistical learning guarantees for third-person imitation from offline observation.
result Strong performance guarantees for transferred policies in the offline setting.
Paper establishes statistical inference for performative predictions.
problem Dynamic influence of predictions on their targets.
method End-to-end framework for estimation and inference under performativity.
result Established central limit theorem for performative settings.
New algorithm improves best arm identification in Bayesian settings.
problem Finding the arm with the highest mean in unknown distributions.
method Developed a variant of successive elimination algorithm.
result Achieved optimal performance in Bayesian setting with logarithmic gap.