arXiv research

Regularizing for or against class selectivity in DNNs improves test accuracy.

Class selectivity affects robustness to corruptions but not to adversarial attacks.

Lower class selectivity makes networks more robust to natural perturbations but more vulnerable to adversarial attacks.

Introduces model class selection to find sets of near-optimal models.

Paper proposes a method to select base classes for few-shot learning.

Adaptive model selection for RL with unknown function classes.

A new data augmentation method selects mixed classes based on class distances for better performance.

In this paper, a novel feature selection method is presented, which is based on Class-Separability (CS) strategy and Data Envelopment Analysis (DEA). To better capture the relationship between features and the class, class labels are separated into individual variables and relevance and redundancy are explicitly handle…

New algorithm offers costless model selection in contextual bandits.

Despite being very effective in several classification tasks, Dynamic Ensemble Selection (DES) techniques can select classifiers that classify all samples in the region of competence as being from the same class. The Frienemy Indecision REgion DES (FIRE-DES) tackles this problem by pre-selecting classifiers that correc…

Random Forest variable importance is improved by class balancing techniques.

A novel method selects genes for high-dimensional gene expression data with class imbalance.

We analyze general model selection procedures using penalized empirical loss minimization under computational constraints. While classical model selection approaches do not consider computational aspects of performing model selection, we argue that any practical model selection procedure must not only trade off estimat…

Paper introduces a new performance metric for class imbalance datasets.

Anomaly detection based on one-class classification algorithms is broadly used in many applied domains like image processing (e.g. detection of whether a patient is "cancerous" or "healthy" from mammography image), network intrusion detection, etc. Performance of an anomaly detection algorithm crucially depends on a ke…

Selective classification improves trading strategies by abstaining from predictions.

A system of nested dichotomies is a method of decomposing a multi-class problem into a collection of binary problems. Such a system recursively splits the set of classes into two subsets, and trains a binary classifier to distinguish between each subset. Even though ensembles of nested dichotomies with random structure…

DSAEE FS selects features for imbalanced data.

Class-imbalance refers to classification problems in which many more instances are available for certain classes than for others. Such imbalanced datasets require special attention because traditional classifiers generally favor the majority class which has a large number of instances. Ensemble of classifiers have been…

A new algorithm selects models for contextual bandits, reducing regret.

The study explores whether model selection guarantees apply to contextual bandits.

Class-imbalance refers to classification problems in which many more instances are available for certain classes than for others. Such imbalanced datasets require special attention because traditional classifiers generally favor the majority class which has a large number of instances. Ensemble of classifiers have been…

Study model selection in batch policy optimization with three error sources.

New method proves neural networks can select features consistently.

This paper proposes a method for multi-class classification problems, where the number of classes K is large. The method, referred to as Candidates vs. Noises Estimation (CANE), selects a small subset of candidate classes and samples the remaining classes. We show that CANE is always consistent and computationally effi…

An ensemble technique is characterized by the mechanism that generates the components and by the mechanism that combines them. A common way to achieve the consensus is to enable each component to equally participate in the aggregation process. A problem with this approach is that poor components are likely to negativel…

Feature selection is one of the most fundamental problems in machine learning. An extensive body of work on information-theoretic feature selection exists which is based on maximizing mutual information between subsets of features and class labels. Practical methods are forced to rely on approximations due to the diffi…

New algorithms for model selection in linear bandits adapt to instance complexity.

Selection of input features such as relevant pieces of text has become a common technique of highlighting how complex neural predictors operate. The selection can be optimized post-hoc for trained models or incorporated directly into the method itself (self-explaining). However, an overall selection does not properly c…

We provide a differentially private algorithm for hypothesis selection. Given samples from an unknown probability distribution $P$ and a set of $m$ probability distributions $\mathcal{H}$, the goal is to output, in a $\varepsilon$-differentially private manner, a distribution from $\mathcal{H}$ whose total variation di…

Tree tensor networks balance model complexity and empirical risk for high-dimensional function approximation.

Conventional mutual information (MI) based feature selection (FS) methods are unable to handle heterogeneous feature subset selection properly because of data format differences or estimation methods of MI between feature subset and class label. A way to solve this problem is feature transformation (FT). In this study,…

Dealing with uncertainty in Bayesian Network structures using maximum a posteriori (MAP) estimation or Bayesian Model Averaging (BMA) is often intractable due to the superexponential number of possible directed, acyclic graphs. When the prior is decomposable, two classes of graphs where efficient learning can take plac…

XNB classifier improves model interpretability by selecting class-specific features.

Feature selection is a crucial preprocessing step in data analytics and machine learning. Classical feature selection algorithms select features based on the correlations between predictive features and the class variable and do not attempt to capture causal relationships between them. It has been shown that the knowle…

A new framework for selecting base classes in multi-class classification boosts accuracy.

For many interesting tasks, such as medical diagnosis and web page classification, a learner only has access to some positively labeled examples and many unlabeled examples. Learning from this type of data requires making assumptions about the true distribution of the classes and/or the mechanism that was used to selec…

A natural way of handling imbalanced data is to attempt to equalise the class frequencies and train the classifier of choice on balanced data. For two-class imbalanced problems, the classification success is typically measured by the geometric mean (GM) of the true positive and true negative rates. Here we prove that G…

Proposes novel wSVMs for sparse learning and accurate probability estimation.

Proposes a new tree-based algorithm for class-imbalanced data.

New gene selection method improves tumor classification accuracy.

This work improves policy evaluation and selection using logarithmic smoothing for pessimistic off-policy estimation.

ReLEX algorithm improves RL efficiency by selecting optimal representations.

The class of chain event graph models is a generalisation of the class of discrete Bayesian networks, retaining most of the structural advantages of the Bayesian network for model interrogation, propagation and learning, while more naturally encoding asymmetric state spaces and the order in which events happen. In this…

Unsupervised clustering can reproduce categorization systems if features and metrics are correctly selected.

The Minimum Description Length (MDL) principle selects the model that has the shortest code for data plus model. We show that for a countable class of models, MDL predictions are close to the true distribution in a strong sense. The result is completely general. No independence, ergodicity, stationarity, identifiabilit…

We introduce the problem of model selection for contextual bandits, where a learner must adapt to the complexity of the optimal policy while balancing exploration and exploitation. Our main result is a new model selection guarantee for linear contextual bandits. We work in the stochastic realizable setting with a seque…

FoLDTree improves oblique decision trees with ULDA, enhancing accuracy and feature selection.