arXiv research

The paper tackles sampling biases by ensuring minority groups are adequately represented in training data.

Biased sampling and missing data complicates statistical problems ranging from causal inference to reinforcement learning. We often correct for biased sampling of summary statistics with matching methods and importance weighting. In this paper, we study nearest neighbor matching (NNM), which makes estimates of populati…

In many applications, different populations are compared using data that are sampled in a biased manner. Under sampling biases, standard methods that estimate the difference between the population means yield unreliable inferences. Here we develop an inference method that is resilient to sampling biases and is able to …

New method identifies parameters of wider shallow neural networks with biases.

The study addresses biases in evaluating molecular optimization methods and proposes methods to reduce these biases.

A new strategy for identifying the best arm in Gaussian bandits with improved exploration.

Approximate Markov chain Monte Carlo (MCMC) offers the promise of more rapid sampling at the cost of more biased inference. Since standard MCMC diagnostics fail to detect these biases, researchers have developed computable Stein discrepancy measures that provably determine the convergence of a sample to its target dist…

SA-GFN corrects biases in GFlowNets due to graph symmetries.

Proposes methods to learn from biased samples, ensuring robust decision rules.

Study reveals biases in ImageNet models are not sufficient for generalization.

We study the convergence rate of stochastic optimization of exact (NP-hard) objectives, for which only biased estimates of the gradient are available. We motivate this problem in the context of learning the structure and parameters of Ising models. We first provide a convergence-rate analysis of deterministic errors fo…

This paper analyzes error bounds for biased SMC samplers in conditional sampling.

New method estimates nested expectations with biased and antithetic sampling.

Theoretical analysis of CNNs' inductive biases and their efficiency in approximating functions.

Resampling outperforms reweighting for correcting biased data in machine learning models.

This study examines biases in flow matching samplers using finite-sample estimation.

We consider adversarial examples for image classification in the black-box decision-based setting. Here, an attacker cannot access confidence scores, but only the final label. Most attacks for this scenario are either unreliable or inefficient. Focusing on the latter, we show that a specific class of attacks, Boundary …

Many machine learning algorithms are trained and evaluated by splitting data from a single source into training and test sets. While such focus on in-distribution learning scenarios has led to interesting advancement, it has not been able to tell if models are relying on dataset biases as shortcuts for successful predi…

The paper improves sparse Gaussian processes by optimizing predictive loss.

To improve the efficiency of Monte Carlo estimation, practitioners are turning to biased Markov chain Monte Carlo procedures that trade off asymptotic exactness for computational speed. The reasoning is sound: a reduction in variance due to more rapid sampling can outweigh the bias introduced. However, the inexactness …

Improved text generation with constraints using discrete auto-regressive biasing.

Method trains a debiased model from a biased one by focusing on samples that contradict the bias.

We introduce biased gradient oracles to capture a setting where the function measurements have an estimation error that can be controlled through a batch size parameter. Our proposed oracles are appealing in several practical contexts, for instance, risk measure estimation from a batch of independent and identically di…

New method improves model robustness to biased data.

New method learns collective variables using autoencoders for molecular simulations.

Balance corrects biased survey data for more accurate insights.

New algorithm finds unbiased subnetworks in biased datasets.

Paper compares Bayesian and de-biased estimators for low-rank matrix completion.

Improves likelihood-free inference by using a new sampling approach to avoid biased data collection.

This work improves surrogate models for balancing accuracy and cost in multi-fidelity methods.

Contrastive Divergence (CD) and Persistent Contrastive Divergence (PCD) are popular methods for training the weights of Restricted Boltzmann Machines. However, both methods use an approximate method for sampling from the model distribution. As a side effect, these approximations yield significantly different biases and…

Self-attention prefers sparse functions of input sequences, reducing sample complexity.

The paper proposes methods to estimate MCMC quality with couplings, bounding Wasserstein distance.

Conditional stochastic optimization covers a variety of applications ranging from invariant learning and causal inference to meta-learning. However, constructing unbiased gradient estimators for such problems is challenging due to the composition structure. As an alternative, we propose a biased stochastic gradient des…

Chinchilla Approach 2 biases neural scaling law estimates, leading to unnecessary compute costs.

With the deluge of digitized information in the Big Data era, massive datasets are becoming increasingly available for learning predictive models. However, in many practical situations, the poor control of the data acquisition processes may naturally jeopardize the outputs of machine learning algorithms, and selection …

Network embedding algorithms are able to learn latent feature representations of nodes, transforming networks into lower dimensional vector representations. Typical key applications, which have effectively been addressed using network embeddings, include link prediction, multilabel classification and community detectio…

New causal models perform poorly when evaluated on biased training sets.

Proposes unbiased estimators for training mixture of experts models.

Improved LIME robustness against adversarial manipulation.

Bayesian optimization improved for biased data.

Active learning suffers from biased non-response, which this paper addresses.

In the early history of positive-unlabeled (PU) learning, the sample selection approach, which heuristically selects negative (N) data from U data, was explored extensively. However, this approach was later dominated by the importance reweighting approach, which carefully treats all U data as N data. May there be a new…

Model proposes neural network for continuous time dynamics with inductive biases.

In this paper, we propose a new framework for mitigating biases in machine learning systems. The problem of the existing mitigation approaches is that they are model-oriented in the sense that they focus on tuning the training algorithms to produce fair results, while overlooking the fact that the training data can its…

A self-supervised debiasing method using rank regularization mitigates spurious correlations in neural networks.

Enhanced diffusion sampling tackles rare event sampling in biomolecular simulations.

PAC-MCTS addresses biased search in LLM-guided planning by dynamically pruning.