arXiv research

Multi-expert L2D underfits more severely, requiring new methods.

We present an intriguing question about lattice points in triangles where Pick's formula is "almost correct". The question has its origin in knot theory, but its statement is purely combinatorial. After more than 30 years the topological question was recently solved, but the lattice point problem is still open.

Sampling more can make models more confident in wrong answers, not better.

The paper ranks experts based on task performance with noisy evaluations.

Study on learning to defer to multiple experts with consistent surrogates and confidence calibration.

The paper combines Bitcoin price models with expert corrections for better predictions.

LLMs struggle with zero-shot annotation tasks due to model-internalized priors.

A2A metric evaluates bias correction methods, reducing ATE estimation errors.

Cryo-electron microscopy (cryo-EM) studies using single particle reconstruction are extensively used to reveal structural information on macromolecular complexes. Aiming at the highest achievable resolution, state of the art electron microscopes automatically acquire thousands of high-quality micrographs. Particles are…

The study examines methods to correct measurement error in nutritional epidemiology studies.

A deep learning model corrects precipitation bias without expert knowledge.

We consider a basic problem at the interface of two fundamental fields: submodular optimization and online learning. In the online unconstrained submodular maximization (online USM) problem, there is a universe $[n]=\{1,2,...,n\}$ and a sequence of $T$ nonnegative (not necessarily monotone) submodular functions arrive …

Improves reward bounds for prediction with expert advice using abstention.

We present a novel approach for supervised domain adaptation that is based upon the probabilistic framework of Gaussian processes (GPs). Specifically, we introduce domain-specific GPs as local experts for facial expression classification from face images. The adaptation of the classifier is facilitated in probabilistic…

Improves accuracy and fairness in prediction systems with multiple domain experts.

New method improves model calibration by adjusting confidence based on prediction correctness.

A method uses confidence scores to handle noisy labels for each instance.

Confidence intervals improve evaluation of binary prediction rules in data mining.

L2D-CD learns to defer expert recommendations in causal discovery.

In the field of reinforcement learning there has been recent progress towards safety and high-confidence bounds on policy performance. However, to our knowledge, no practical methods exist for determining high-confidence policy performance bounds in the inverse reinforcement learning setting---where the true reward fun…

AI-generated variables bias regression estimates; methods correct for invalid inference.

In training speech recognition systems, labeling audio clips can be expensive, and not all data is equally valuable. Active learning aims to label only the most informative samples to reduce cost. For speech recognition, confidence scores and other likelihood-based active learning methods have been shown to be effectiv…

A framework uses a mixture of predictors for semi-supervised inference.

Proposes a Varying-Coefficient MoE model for analyzing dynamic data.

Adversarial MoE learns category-specific models for product search.

Survey of algorithms to correct past mistakes in prediction.

A new L2D system produces calibrated probabilities of expert correctness without sacrificing accuracy.

Paper tackles weakly supervised learning from similarity-confidence data.

Improved AI lung ultrasound segmentation using expert confidence values.

We consider the problem of contextual bandits with stochastic experts, which is a variation of the traditional stochastic contextual bandit with experts problem. In our problem setting, we assume access to a class of stochastic experts, where each expert is a conditional distribution over the arms given a context. We p…

Paper proposes a method to improve deep neural networks' confidence estimates.

New method uses variational inference to handle confounding in imitation learning.

The paper addresses the gap between theoretical and practical confidence set widths in universal inference.

Proposes unbiased estimators for training mixture of experts models.

We develop the setting of sequential prediction based on shifting experts and on a "smooth" version of the method of specialized experts. To aggregate experts predictions, we use the AdaHedge algorithm, which is a version of the Hedge algorithm with adaptive learning rate, and extend it by the meta-algorithm Fixed Shar…

New method tackles label noise on imbalanced datasets by considering class-specific uncertainty.

Combines expert knowledge and data for efficient probability distribution inference.

The medical field stands to see significant benefits from the recent advances in deep learning. Knowing the uncertainty in the decision made by any machine learning algorithm is of utmost importance for medical practitioners. This study demonstrates the utility of using Bayesian LSTMs for classification of medical time…

Classifies Calabi hypersurfaces with parallel Fubini-Pick form.

Paper proves a discrete Schwarz-Pick lemma for generalized circle packings.

A contextual care protocol is used by a medical practitioner for patient healthcare, given the context or situation that the specified patient is in. This paper proposes a method to build an automated self-adapting protocol which can help make relevant, early decisions for effective healthcare delivery. The hybrid mode…

Near-optimal per-action regret bounds for sleeping bandits are derived.

Meta-algorithm optimizes nonstochastic bandits with infinitely many experts.

Paper addresses hypothesis space misspecification in learning from human demonstrations and corrections.

Develops a novel fast bootstrap for dependent data with higher-order accuracy.

Forest-guided smoothing uses random forest outputs for interpretable local smoothers.

New method corrects skewed confidence for PbN classification.

Proposes a method to compute valid lower confidence bounds for multiple models selected based on their performance.