New method corrects selection bias in post-selective inference for Group LASSO.
problem Inference after Group LASSO selection is unreliable.
method Develops a consistent, post-selective Bayesian method to adjust for selection bias.
result Corrects bias in recovering effects of selected variables.
PS-DME evaluates model performance and reliability after data-dependent selection.
problem Evaluating model performance and reliability when data is used for selection and evaluation.
method Post-selection distributional model evaluation (PS-DME) using e-values to control false coverage rate.
result PS-DME provides reliable comparison of model configurations across different reliability levels.
The paper discusses methods for interval estimation of coefficients in penalized regression models for insurance data.
problem Valid inference on coefficients after feature selection in GLM family for insurance data.
method Proposes methodologies for constructing confidence intervals of coefficients after feature selection in GLM family.
result Valid inference on coefficients after feature selection in GLM family for insurance data.
Paper improves Lasso for S&P500 index tracking with post-selection inference.
problem Index tracking for S&P500 with many applications.
method Used Lasso for dimension reduction and post-selection inference.
result Lasso method for S&P500 index tracking shows high performance.
Study examines inference methods after variable selection in Cox models.
problem Bias and misleading inference after variable selection in Cox models.
method Simulation study of inference procedures for Lasso and adaptive Lasso in Cox models.
result Performance of inference procedures varies, with debiased Lasso showing promise.
Paper proposes a new method combining random forests and Lasso selection.
problem Improving random forest performance by applying Lasso regression.
method Adaptive Lasso weighting applied to random forest predictions.
result Unified framework strictly outperforms other methods in simulations and real-world datasets.
We develop a general approach to valid inference after model selection. At the core of our framework is a result that characterizes the distribution of a post-selection estimator conditioned on the selection event. We specialize the approach to model selection by the lasso to form valid confidence intervals for the sel…
Develops methods to adjust prediction set coverage based on post-selection analysis.
problem Adjusting prediction set coverage after initial analysis to better fit specific needs.
method Post-selection conformal inference to adjust miscoverage levels.
result Allows for trade-off between coverage and prediction set quality.
New framework for valid hypothesis testing in complex data settings.
problem Challenges in classical hypothesis testing frameworks.
method Add and subtract external noise to partition data, orthogonalize, and test hypotheses.
result Valid hypothesis tests can be conducted under minimal assumptions.
Proposes HSIC-Lasso for selective inference in non-linear data.
problem Detecting influential features in non-linear and high-dimensional data.
method Model-free HSIC-Lasso based on truncated Gaussians and polyhedral lemma.
result Tight control of type-I error even for small sample sizes.
While statistics and machine learning offers numerous methods for ensuring generalization, these methods often fail in the presence of adaptivity---the common practice in which the choice of analysis depends on previous interactions with the same dataset. A recent line of work has introduced powerful, general purpose a…
"Which Generative Adversarial Networks (GANs) generates the most plausible images?" has been a frequently asked question among researchers. To address this problem, we first propose an \emph{incomplete} U-statistics estimate of maximum mean discrepancy MMDinc to measure the distribution discrepancy betwee…
Paper simplifies data carving inference with a parametric distribution.
problem Valid inference after selection with data carving.
method Developed a parametric distribution for data carving inference.
result Exact inference for data carving can be computed trivially.
We propose a novel kernel based post selection inference (PSI) algorithm, which can not only handle non-linearity in data but also structured output such as multi-dimensional and multi-label outputs. Specifically, we develop a PSI algorithm for independence measures, and propose the Hilbert-Schmidt Independence Criteri…
Estimates true Sharpe ratio of selected assets with various methods.
problem Estimating the true Sharpe ratio of a selected asset with high in-sample ratio.
method Polyhedral lemma, James Stein shrinkage, debiasing, thresholding, empirical Bayes.
result James Stein estimator performs best across various parameter values.
New algorithms for sampling in constrained domains without learning rates.
problem Sampling in constrained domains with fairness constraints and post-selection inference.
method Coin betting ideas from convex optimisation and a unifying framework for constrained sampling.
result Our algorithms achieve competitive performance without hyperparameter tuning.
A method to split a data point into two parts that individually cannot reconstruct the whole, but together can.
problem Splitting a single data point into two parts such that neither can reconstruct the whole but together can.
method Borrowing ideas from Bayesian inference to achieve a continuous analog of data splitting.
result A method to achieve data fission, enabling post-selection inference in finite samples.
Refining one's hypotheses in the light of data is a common scientific practice; however, the dependency on the data introduces selection bias and can lead to specious statistical analysis. An approach for addressing this is via conditioning on the selection procedure to account for how we have used the data to generate…
New samplers minimize KL divergence for constrained and non-Euclidean geometries.
problem Efficient sampling from constrained and non-Euclidean distributions.
method Stein Variational Mirror Descent and Mirrored Stein Variational Gradient Descent.
result New samplers converge more rapidly and accurately than prior methods.
Measuring divergence between two distributions is essential in machine learning and statistics and has various applications including binary classification, change point detection, and two-sample test. Furthermore, in the era of big data, designing divergence measure that is interpretable and can handle high-dimensiona…
New framework for cyclic quantum causal models with graph separation property.
problem Understanding causal relationships in feedback processes and exotic scenarios.
method Introducing a robust probability rule and a novel graph-separation property, p-separation.
result Established graph-separation properties for all consistent cyclic causal models.
Proposes MinPEN framework for estimating relationships in multivariate models.
problem Estimating relationships between multivariate outcomes in statistical learning.
method MinPEN framework using minimum function penalty for non-convex optimization.
result Theoretical and practical validation of MinPEN framework for multivariate models.
New method splits unknown covariance Gaussians into independent parts.
problem Splitting multivariate Gaussian data with unknown covariance.
method Developed a general algorithm for decomposing unknown covariance Gaussians.
result Demonstrated decomposition for single multivariate Gaussian with unknown covariance.
New method for estimating high-dimensional binary time series coefficients.
problem Statistical inference for high-dimensional binary time series.
method Post-selection estimator and second-order wild bootstrap algorithm.
result Good finite-sample performance of the proposed method.
Develops a new framework for causal models on cyclic graphs, solving unique solvability issues.
problem Challenges in specifying unique probability distributions for cyclic functional causal models.
method Introduces a new probability rule and graph-separation property (p-separation) for cyclic fCMs.
result Proves p-separation is sound and complete for all consistent cyclic fCMs, recovering d-separation for DAGs.
Finding statistically significant high-order interaction features in predictive modeling is important but challenging task. The difficulty lies in the fact that, for a recent applications with high-dimensional covariates, the number of possible high-order interaction features would be extremely large. Identifying stati…
Due to the increasing availability of high-dimensional empirical applications in many research disciplines, valid simultaneous inference becomes more and more important. For instance, high-dimensional settings might arise in economic studies due to very rich data sets with many potential covariates or in the analysis o…
We address the problem of non-parametric multiple model comparison: given l candidate models, decide whether each candidate is as good as the best one(s) or worse than it. We propose two statistical tests, each controlling a different notion of decision errors. The first test, building on the post selection inference…
We propose a statistical inference framework for the component-wise functional gradient descent algorithm (CFGD) under normality assumption for model errors, also known as L2-Boosting. The CFGD is one of the most versatile tools to analyze data, because it scales well to high-dimensional data sets, allows for a very…
Study on estimating causal effects with limited data and multiple environments.
problem Estimating causal effects under hidden confounding with unpaired data and sparse effects.
method Instrumental variable (IV) regression with cross-fold sample splitting and ℓ1-regularized estimation. result Proposed GMM-type estimator is consistent as the number of environments grows.
Research shows bias in machine learning can be due to algorithmic flaws, not just data.
problem Underestimation bias in machine learning algorithms.
method Initial research to understand factors contributing to bias in classification algorithms.
result Regularization methods to address overfitting can also accentuate bias.
Ever since the proof of asymptotic normality of maximum likelihood estimator by Cramer (1946), it has been understood that a basic technique of the Taylor series expansion suffices for asymptotics of M-estimators with smooth/differentiable loss function. Although the Taylor series expansion is a purely deterministic …
Financial LLMs need explicit bias consideration to avoid invalid results.
problem Finance-specific biases inflate performance and contaminate backtests.
method Identified five recurring biases and proposed a Structural Validity Framework.
result Explicit bias consideration is necessary for valid deployment claims.
We study the phenomenon of bias amplification in classifiers, wherein a machine learning model learns to predict classes with a greater disparity than the underlying ground truth. We demonstrate that bias amplification can arise via an inductive bias in gradient descent methods that results in the overestimation of the…
It has been noticed that some external CVIs exhibit a preferential bias towards a larger or smaller number of clusters which is monotonic (directly or inversely) in the number of clusters in candidate partitions. This type of bias is caused by the functional form of the CVI model. For example, the popular Rand index (R…
Mitigates gender bias amplification in model predictions.
problem Gender bias amplification in model predictions.
method Posterior regularization to mitigate bias.
result Almost removes gender bias amplification in model predictions.
Depth uncertainty networks don't improve with bias correction, contrary to expectations.
problem Improving performance in active learning with overparameterised models like NNs.
method Depth uncertainty networks, compared to underparameterised models, show no improvement in performance with bias correction.
result Depth uncertainty networks do not improve with bias correction, unlike underparameterised models.
We quantify causal bias in continuous treatment settings.
problem Identifying and quantifying causal bias in continuous treatment scenarios.
method Developed a novel characterization of causal bias in structural causal models, proving conditions for zero bias and efficient estimation.
result Causal bias can be estimated efficiently under certain structural equation restrictions, allowing for causal regularization of predictive models.
Selective inference for group lasso estimators across various distributions and covariates.
problem Developing selective inference methods for group lasso estimators.
method Randomized group-regularized optimization problem with post-selection likelihood.
result Selective point estimator and Wald-type confidence regions for regression parameters.
We study the convergence of the predictive surface of regression trees and forests. To support our analysis we introduce a notion of adaptive concentration for regression trees. This approach breaks tree training into a model selection phase in which we pick the tree splits, followed by a model fitting phase where we f…
The paper introduces Relative Bias to quantify LLM bias systematically.
problem Quantifying bias in LLMs is challenging due to ambiguity and rapid model emergence.
method Relative Bias framework using Embedding Transformation and LLM-as-a-Judge methodologies.
result The two scoring methods show strong alignment, providing a systematic approach.
Interpolated-MLPs control inductive bias for better performance in low-compute tasks.
problem Low-compute performance gap between MLPs and CNNs.
method Introduced Interpolated MLP (I-MLP) approach to control inductive bias incrementally.
result Continuous logarithmic relationship between inductive bias and performance in low-compute tasks.
SSMs have a built-in bias towards low-frequency components, which can be adjusted.
problem Frequency bias in SSMs affects their performance on long-range sequences.
method Proposed two mechanisms to tune frequency bias: scaling initialization or applying a Sobolev-norm-based filter.
result Tuning frequency bias improves SSMs' performance on long-range sequence learning tasks.
A bias classifier is introduced to resist adversarial attacks.
problem Resisting adversarial attacks on deep neural networks (DNNs).
method Introducing the bias part of a DNN with Relu as the activation function as a classifier, and adding a random first-degree part to make it information-theoretically safe.
result The bias classifier is more robust than DNNs of similar size against adversarial attacks.
UBM transfers bias mitigation from upstream to downstream tasks efficiently.
problem Bias in fine-tuned language models across various tasks.
method Apply bias mitigation to an upstream model, then fine-tune a downstream model on this mitigated model.
result UBM effects transfer to new downstream tasks, creating less biased models.
Social bias in machine learning has drawn significant attention, with work ranging from demonstrations of bias in a multitude of applications, curating definitions of fairness for different contexts, to developing algorithms to mitigate bias. In natural language processing, gender bias has been shown to exist in contex…
Active learning introduces bias; this paper fixes it.
problem Bias in active learning due to non-representative training data.
method Formalized bias, identified situations where it's harmful/helpful, introduced corrective weights.
result Corrective weights can improve active learning, especially with overparameterized models.
Bias correction improves language model training performance.
problem Stochastic update bias in preconditioned optimizers.
method Cross-fitted preconditioning and variance-corrected inversion.
result Reduces held-out pretraining loss by 0.15 nats.