Exposure bias has been regarded as a central problem for auto-regressive language models (LM). It claims that teacher forcing would cause the test-time generation to be incrementally distorted due to the training-generation discrepancy. Although a lot of algorithms have been proposed to avoid teacher forcing and theref…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,742 papers · 148 categories
Trend · papers per month
48 results for “exposure bias”
New method debiases selection bias in PU classification with exposure data.
problem Binary classification from positive and unlabeled data with selection bias.
method Automatic Debiased PUE (ADPUE) learning method.
result ADPUE outperforms traditional PU learning methods on various datasets.
The paper studies and mitigates bias in recommender systems.
problem Bias in recommender systems caused by user feedback loops.
method Modeling user exposure and proposing debiasing strategies.
result Recommender systems are biased and depend on user exposure.
Paper tackles exposure bias in recommender systems using contrastive learning.
problem Exposure bias in large-scale recommender systems.
method Contrastive learning to reduce exposure bias via inverse propensity weighting.
result Contrastive learning effectively reduces exposure bias in recommender systems.
In epidemiology, identifying the effect of exposure variables in relation to a time-to-event outcome is a classical research area of practical importance. Incorporating propensity score in the Cox regression model, as a measure to control for confounding, has certain advantages when outcome is rare. However, in situati…
The paper tackles robust design selection for online experiments under uncertain interference mechanisms.
problem Designing experiments in ads, recommendations, and member-experience systems when interference mechanisms are unknown.
method Formulates the problem as robust design selection over uncertain exposure mechanisms. Compares designs by worst-case planning risk over an ambiguity set combining various factors.
result Develops a geometry-aware guarantee and robust selector theorem with excess-risk control, exact recovery under separation, and certified shortlists when the risk surface is flat.
Exposure bias refers to the train-test discrepancy that seemingly arises when an autoregressive generative model uses only ground-truth contexts at training time but generated ones at test time. We separate the contributions of the model and the learning framework to clarify the debate on consequences and review propos…
The paper tackles biases in session-based recommender systems by modeling user interest as a stochastic process.
problem Data uncertainty, popularity bias, and exposure bias in session-based recommender systems.
method The paper proposes treating user interest as a stochastic process in the latent space, debiasing item embeddings, modeling dense user interest, and introducing fake targets to simulate extended exposure.
result The proposed approach mitigates challenges in session-based recommender systems, as shown by computational experiments on various datasets.
The study examines methods to correct measurement error in nutritional epidemiology studies.
problem Measurement error in nutritional studies leads to biased and underconfident estimates.
method The article reviews various bias-correction models for exposure variables in nutritional epidemiology.
result Bias-correction methods are essential for accurate inference in nutritional studies.
Modeling incentives for content creators on algorithm-curated platforms.
problem Maximizing exposure for content creators on algorithmic platforms.
method Formalized exposure game model, proving effects of algorithmic choices on equilibria, proposing tools for finding equilibria.
result Algorithmic choices significantly affect content exposure and creator behavior.
Multi-label classification (MLC) assigns multiple labels to each sample. Prior studies show that MLC can be transformed to a sequence prediction problem with a recurrent neural network (RNN) decoder to model the label dependency. However, training a RNN decoder requires a predefined order of labels, which is not direct…
A new method improves text generation quality and diversity.
problem Exposure bias in Maximum Likelihood Estimation for text generation.
method ψ-MLE, a new training scheme based on density ratio estimation.
result ψ-MLE outperforms Maximum Likelihood Estimation and other models in text generation quality and diversity.
Observational cohort studies with oversampled exposed subjects are typically implemented to understand the causal effect of a rare exposure. Because the distribution of exposed subjects in the sample differs from the source population, estimation of a propensity score function (i.e., probability of exposure given basel…
TMLE improves causal effect estimation in missing data scenarios with various positivity violations.
problem Estimating causal effects in studies with missing data and positivity violations.
method Targeted Maximum Likelihood Estimation (TMLE) with various missing data methods.
result Complete cases with TMLE incorporating an outcome-missingness model exhibit lower bias and greater robustness against positivity violations.
Estimates treatment effects in bipartite systems with partial eligibility and interference.
problem Randomized experiments in bipartite systems with partial treatment eligibility and interference.
method Formalizes eligibility-constrained bipartite experiments, defines PTTE and STTE, identifies conditions, develops ensemble estimators, introduces projection.
result Proposed estimators recover PTTE and STTE with low bias and variance, corrects interference bias in field experiments.
New proof links initial class bias to DNN trainability, challenging traditional understanding.
problem Understanding the initial class bias in DNNs and its impact on trainability.
method Theoretical proof linking initial class bias to mean field theories of DNNs.
result Efficient learning is connected to a network's prejudice towards a specific class, contradicting traditional understanding.
Develops sparse portfolio strategy for high-dimensional assets.
problem Sparse wealth allocations in high dimensions are limited by existing approaches.
method Establishes theoretical bounds and empirical analysis of sparse weight estimators.
result Sparse portfolios are robust to recessions and can be used as a hedging vehicle.
Second Order Riskq-fin.PM
Managing a portfolio to a risk model can tilt the portfolio toward weaknesses of the model. As a result, the optimized portfolio acquires downside exposure to uncertainty in the model itself, what we call "second order risk." We propose a risk measure that accounts for this bias. Studies of real portfolios, in asset-by…
This research develops a new framework to measure AI investment returns considering both gains and risks.
problem Traditional ROI calculations fail to account for AI's dual impact on risk reduction and new exposures.
method Integrates ISO 42001 and regulatory exposure into a comprehensive financial framework using risk quantification methods.
result Accurate AI investment evaluation requires modeling both productivity gains and risk exposures.
Model improves CVR estimation in recommender systems by mitigating bias and overlooking causal relationships.
problem Data sparsity and sample selection bias in CVR estimation.
method Entire Space Counterfactual Multitask Model (ESCM) incorporating counterfactual risk minimizer.
result Significantly enhances recommendation performance by effectively mitigating bias and overlooking causal relationships.
While implicit feedback (e.g., clicks, dwell times, etc.) is an abundant and attractive source of data for learning to rank, it can produce unfair ranking policies for both exogenous and endogenous reasons. Exogenous reasons typically manifest themselves as biases in the training data, which then get reflected in the l…
The study corrects measurement error in evaluating health effects of multiple pollutants.
problem Bias in estimating health effects of air pollution constituents due to mismeasurement.
method Used a linear regression calibration model and extended DML approach to correct for measurement error.
result Identified two PM2.5 constituents (Br and Mn) that show a negative causal effect on cognitive function after correction.
Measurement error in observational datasets can lead to systematic bias in inferences based on these datasets. As studies based on observational data are increasingly used to inform decisions with real-world impact, it is critical that we develop a robust set of techniques for analyzing and adjusting for these biases. …
Green stocks show less factor exposure heterogeneity compared to brown stocks.
problem Exploring differences in factor exposure between green and brown stocks.
method Examined S&P 500 firms grouped by greenhouse gas emissions, analyzing factor exposure over 2014-2020.
result Green stocks have less factor exposure heterogeneity than brown stocks, except for the value factor.
A deep learning approach for computations of exposure profiles for high-dimensional Bermudan optionsq-fin.CP
Deep learning approximates Bermudan option exposures and future values.
problem Computing accurate expected and future exposures for high-dimensional Bermudan options.
method Neural network-based approach combining Deep Optimal Stopping and regression.
result Neural network approximations of pathwise option values are more accurate.
Study shows short exposure and systematic risk exposure affect disposition effect asymmetries.
problem Understanding disposition effect in short vs long exposure positions and systematic risk.
method Generalized Odean measures, introduced Value metric, implemented dispositionEffect R package.
result Short positions exhibit weaker disposition effect than long positions under narrow framing, reversing in integrated framing.
Randomized neural networks improve exposure and CVA estimation for American options.
problem Estimation of exposure and CVA for American options
method Randomized neural networks
result Improves convergence and efficiency in high-dimensional problems
Data mining and machine learning techniques such as classification and regression trees (CART) represent a promising alternative to conventional logistic regression for propensity score estimation. Whereas incomplete data preclude the fitting of a logistic regression on all subjects, CART is appealing in part because s…
Enhanced X-ray polarimetry with deep learning for better exposure times.
problem Improving sensitivity of X-ray telescopic observations with imaging polarimeters.
method A weighted maximum likelihood combination of predictions from a deep ensemble of ResNet convolutional neural networks trained on Monte Carlo event simulations.
result Improves effective exposure times by ~45% for power-law source spectra.
Valuation of Credit Valuation Adjustment (CVA) has become an important field as its calculation is required in Basel III, issued in 2010, in the wake of the credit crisis. Exposure, which is defined as the potential future loss of a default event without any recovery, is one of the key elementsfor pricing CVA. This pap…
Hidden Markov Models analyze mobile health data to identify APNS states.
problem Subjective self-report measures of APNS lead to errors and biases.
method Exploratory hidden Markov factor models and Stabilized Expectation-Maximization algorithm.
result Identified homogeneous APNS states and dynamic transitions.
Paper optimizes neural networks for Bermudan option pricing with faster convergence and risk management tools.
problem Efficiently pricing Bermudan options with static hedging and risk management.
method Monte-Carlo-based artificial neural network framework with novel optimisation algorithm.
result The proposed neural network accelerates convergence and provides improved risk management tools.
Self-reinforcing feedback loops in personalization systems are typically caused by users choosing from a limited set of alternatives presented systematically based on previous choices. We propose a Bayesian choice model built on Luce axioms that explicitly accounts for users' limited exposure to alternatives. Our model…
We study the impact of central clearing of over-the-counter (OTC) transactions on counterparty exposures in a market with OTC transactions across several asset classes with heterogeneous characteristics. The impact of introducing a central counterparty (CCP) on expected interdealer exposure is determined by the tradeof…
The paper proposes using function approximations to reduce the computational burden in measuring counterparty credit exposure.
problem The need for regular exposure calculations in finance, balancing between computational cost and risk simplification.
method Replacing derivative pricers with function approximations, proving error bounds, and using Chebyshev interpolation for convergence.
result Derives probabilistic and finite sample error bounds, showing significant run-time reductions and asymptotic efficiency gains.
Physen-Noise2Noise tackles defocus deblurring in low-light conditions with physics-guided self-supervised learning.
problem Defocus deblurring in low-light conditions with complex biased noise.
method Physics-guided self-supervised deblurring framework that leverages noisy multi-frame observations and a learnable noise bias parameter.
result Physen-Noise2Noise consistently outperforms state-of-the-art methods in defocus deblurring with complex biased noise.
Study prenatal PM2.5 exposure and 4th grade reading scores, identifying critical windows of susceptibility.
problem Understanding the impact of prenatal PM2.5 exposure on educational outcomes.
method Developed a locally adaptive Bayesian regression model with B-spline basis expansion and dynamic shrinkage priors.
result Prenatal PM2.5 exposure during early and late pregnancy is most adverse for 4th grade reading scores.
A Reinforced Topic-Aware Convolutional Sequence-to-Sequence Model for Abstractive Text Summarizationcs.CL
In this paper, we propose a deep learning approach to tackle the automatic summarization tasks by incorporating topic information into the convolutional sequence-to-sequence (ConvS2S) model and using self-critical sequence training (SCST) for optimization. Through jointly attending to topics and word-level alignment, o…
Mack's estimator improves chain ladder prediction for large exposure insurance models.
problem Uncertainty quantification in compound Poisson loss models.
method Large exposure asymptotics applied to Mack's estimator.
result Chain ladder prediction uncertainty can be quantified without model assumptions.
Study estimates personalized effects of maternal PM2.5 exposure on birth weight.
problem Identify critical windows and heterogeneity in maternal PM2.5 exposure effects on birth weight.
method Heterogeneous Distributed Lag Models and Bayesian Additive Regression Trees.
result Evidence of heterogeneity in PM2.5-birth weight relationship, with some dyads showing 3x larger decrease.
Fast Calculation of Credit Exposures for Barrier and Bermudan options using Chebyshev interpolationq-fin.CP
We introduce a new method to calculate the credit exposure of Bermudan, discretely monitored barrier and European options. Core of the approach is the application of the dynamic Chebyshev method of Glau et al. (2019). The dynamic Chebyshev method delivers a closed form approximation of the option prices along the paths…
Collaborative filtering analyzes user preferences for items (e.g., books, movies, restaurants, academic papers) by exploiting the similarity patterns across users. In implicit feedback settings, all the items, including the ones that a user did not consume, are taken into consideration. But this assumption does not acc…
Despite some empirical success at correcting exposure bias in machine translation, scheduled sampling algorithms suffer from a major drawback: they incorrectly assume that words in the reference translations and in sampled sequences are aligned at each time step. Our new differentiable sampling algorithm addresses this…
BN^2MF identifies unknown exposure patterns in environmental mixtures.
problem Identifying unknown exposure patterns in environmental mixtures.
method Bayesian non-parametric non-negative matrix factorization (BN^2MF) with non-negative continuous priors and a non-parametric sparse prior.
result Estimates patterns of chemical exposures without specifying the number of patterns.
We introduce a new method to calculate the credit exposure of European and path-dependent options. The proposed method is able to calculate accurate expected exposure and potential future exposure profiles under the risk-neutral and the real-world measure. Key advantage of is that it delivers an accuracy comparable to …
Researchers develop a new method to assess variable importance in spatial machine learning models for air pollution exposure prediction.
problem Understanding the mechanism captured by machine learning models in air pollution studies, especially with spatial correlation.
method Leave-one-out approach for variable importance measure applicable to models with separable mean and covariance components.
result The new method highlights differences in model mechanisms even for similar prediction accuracies.
Bayesian method detects outliers and uncertain points in data.
problem Detecting outliers and uncertain points in data using Bayesian methods.
method Generative model of data curation for aleatoric uncertainty, combining with epistemic uncertainty and outlier exposure.
result Principled Bayesian approach outperforms methods using aleatoric or epistemic uncertainty alone.
CENNSurv models cumulative effects of time-dependent exposures on survival outcomes.
problem Challenges in modeling cumulative effects of time-dependent exposures on survival outcomes.
method CENNSurv, a novel deep learning approach that captures dynamic risk relationships from time-dependent data.
result CENNSurv reveals multi-year lagged and short-term behavioral shifts in survival outcomes.