The paper addresses biased preferences in candidate selection, proposing a fair and utility-maximizing algorithm.
problem Selecting candidates for institutions with biased preferences and limited capacities.
method An algorithm that considers group fairness and true utility, proving near-optimal results under distributional assumptions.
result The proposed algorithm achieves near-optimal group fairness and near-maximal true utility, even in biased settings.
Survival trees exhibit end-cut preference, leading to biased splits.
problem End-cut preference in survival trees causes biased splits and poor interpretability.
method Proposed a smooth sigmoid surrogate (SSS) approach to replace hard-threshold indicator function.
result Smooth sigmoid surrogate (SSS) effectively mitigates end-cut preference in survival trees.
Randomly initialized transformers show extreme token preferences.
problem Structural biases in randomly initialized transformers.
method Dissection of transformer architecture at initialization.
result Initialization-induced biases persist throughout training.
A simple method reduces bias in LLM auto-evaluators by controlling output length.
problem Bias in LLM auto-annotators, particularly length bias.
method Simple regression analysis to control for length difference and other mediators.
result Improved robustness and increased correlation with human preferences.
LLM evaluation suffers from systematic biases and lacks reliable positive judgments.
problem LLM evaluation suffers from systematic biases and lacks reliable positive judgments.
method Formulate LLM evaluation as a positive-unlabelled learning problem and propose a geometric auditing framework based on Partial Optimal Transport.
result Improved alignment with human preferences, increased robustness to presentation biases, and interpretable confidence estimates.
No free lunch theorems suggest inductive biases are needed, but we show neural networks prefer low-complexity data.
problem The need for inductive biases in machine learning.
method Analysis of Kolmogorov complexity and neural network behavior on various datasets.
result Neural networks prefer low-complexity data, suggesting inductive biases are not always necessary.
Develops methods to correct bias in AI feedback for more accurate alignment.
problem Systematic bias in AI feedback compared to human labels.
method Two debiased alignment methods: DDPO and DIPO.
result Methods improve alignment efficiency and performance close to human-labeled data.
Inverse reinforcement learning (IRL) enables an agent to learn complex behavior by observing demonstrations from a (near-)optimal policy. The typical assumption is that the learner's goal is to match the teacher's demonstrated behavior. In this paper, we consider the setting where the learner has its own preferences th…
Study finds gender bias in human evaluators and shows how machine learning can mitigate it.
problem Gender bias in human decision-making on micro-lending platforms.
method Structural econometric model and machine learning algorithms trained on real-world data.
result Machine learning algorithms can mitigate both preference-based and belief-based biases.
Analyze SGD with biased gradients, improving convergence rates and accuracy.
problem Analyzing the convergence of SGD with biased gradients.
method Derive convergence results for smooth non-convex functions and quantify the impact of bias magnitude.
result Improved rates under the Polyak-Lojasiewicz condition and insights into how bias magnitude affects accuracy and convergence.
Self-attention prefers sparse functions of input sequences, reducing sample complexity.
problem Understanding the inductive biases of self-attention in modeling long-range dependencies.
method Theoretical analysis and synthetic experiments to probe sample complexity of learning sparse functions with Transformers.
result Bounded-norm Transformer networks can represent sparse functions of the input sequence with logarithmic sample complexity.
Mitigates biases in reward models using variational inference.
problem Spurious correlations in reward models that align large language models with human preferences.
method Formulates data-generating process, identifies non-spurious latent variables, and uses variational inference to recover them.
result Effective mitigation of spurious correlation issues, yielding more robust reward models.
LLMs show biases in investment analysis, leading to unreliable recommendations.
problem LLMs face conflicts between pre-trained knowledge and real-time market data, leading to biases in investment analysis.
method Experimental framework to investigate emergent behaviors in LLMs, analyzing sector, size, and momentum biases.
result Distinct, model-specific biases observed, including a tendency to prefer technology stocks, large-cap stocks, and contrarian strategies.
This paper identifies and analyzes biases in risk-adjusted index weighting methods, affecting social welfare and market fairness.
problem Biases in risk-adjusted index weighting methods lead to tracking errors and fraud in indices and ETFs.
method Characterizes and analyzes the biases and adverse effects of risk-adjusted index weighting methods.
result These biases reduce social welfare and can enable harmful arbitrage activities.
Study reveals DNNs prefer easy-to-learn cues over essential ones in image recognition.
problem DNNs learn easy-to-learn features that aren't essential to the task.
method WCST-ML training setup with shortcut cues on synthetic and face datasets.
result DNNs converge to solutions focusing on preferred cues, leading to flat minima.
New benchmark uncovers hidden biases in LLMs that refuse to answer certain queries.
problem Evaluating fairness in LLMs, especially in sensitive applications.
method Silenced Bias Benchmark (SBB) using activation steering to reduce model refusals during QA.
result Exposes hidden unfair preferences in LLMs' latent space, distinguishing direct responses from underlying fairness issues.
We propose a scalable Bayesian preference learning method for jointly predicting the preferences of individuals as well as the consensus of a crowd from pairwise labels. Peoples' opinions often differ greatly, making it difficult to predict their preferences from small amounts of personal data. Individual biases also m…
Improved probabilistic forecasts using behavioral transformations.
problem Improving accuracy and consistency of probabilistic asset price forecasts.
method Behavioral transformation of fundamental expectations to disentangle sentiment-induced biases.
result Substantial forecast gains across various models and risk-preferences.
The estimation of advantage is crucial for a number of reinforcement learning algorithms, as it directly influences the choices of future paths. In this work, we propose a family of estimates based on the order statistics over the path ensemble, which allows one to flexibly drive the learning process, towards or agains…
The study aims to prevent unfair content presentation in recommender systems.
problem Over- and under-presentation of content leads to biased user preference estimates.
method Two models are considered: one that ignores systematic and limited exposure, and another that conditions on limited exposure.
result Ignoring systematic presentations overestimates promoted options and underestimates censored alternatives.
Meta-Router optimizes LLM selection using gold-standard and preference-based data.
problem Training a high-quality LLM router with combined data sources is challenging due to bias and scarcity.
method Developed an integrative causal router training framework to correct bias and improve routing accuracy.
result Our approach delivers more accurate routing and improves the trade-off between cost and quality.
System learns to combine multiple model components for personalized text generation.
problem Adapting and biasing language models for personal preferences.
method Combines model-defined components, learns activation and probability combination from unlabeled text.
result Directly generates text with personalized components from unlabeled data.
Paper analyzes how present-bias affects carbon emissions and proposes a method to mitigate it.
problem Present-bias impacts carbon emission patterns towards a net zero target.
method Stochastic control techniques adapted from insurance risk theory.
result Higher present-bias leads to excess emissions, and carbon taxes can reduce emissions but beyond a certain point have diminishing returns.
Framework assesses autograders' reliability and biases.
problem Mixed reliability and biases in autograders for LLM evaluation.
method Bayesian GLMs to model evaluation outcomes.
result Explicit quantification of scoring differences and biases.
CausalRM models rewards from user feedback, overcoming noise and bias.
problem Aligning language models with user preferences from noisy, biased feedback.
method Causal-theoretic reward modeling framework addressing noise and bias in observational feedback.
result CausalRM learns accurate reward signals from noisy and biased observational feedback.
The study examines biases in Kiva.org's microfinance platform and proposes methods to mitigate them.
problem Identifying and mitigating hidden biases in lender preferences for different sectors on Kiva.org.
method The study uses causal inference and regression models with Bayesian variable selection methods to investigate and quantify biases. Fairness constraints are then incorporated to maintain comparable results.
result The influence of economic factors and loan attributes on lender preferences varies by sector, and models can be adjusted to mitigate biases.
Paper tackles dueling bandits with delayed feedback, revealing preference bias.
problem Real-world dueling bandit applications often face delays in feedback.
method Introduces biased dueling bandit problem with stochastic delayed feedback, presents two algorithms.
result Two algorithms achieve optimal regret bounds for dueling bandit problems with delay.
RAFT fine-tunes models using high-quality samples to align them with human preferences.
problem Aligning generative models with human ethics and preferences.
method RAFT selects high-quality samples, discards undesired behavior, and fine-tunes the model on filtered samples.
result RAFT improves model performance in reward learning and automated metrics.
Citizen science projects are successful at gathering rich datasets for various applications. However, the data collected by citizen scientists are often biased --- in particular, aligned more with the citizens' preferences than with scientific objectives. We propose the Shift Compensation Network (SCN), an end-to-end l…
Study replicates reference-dependent preferences impact on risk-return trade-off in Chinese stock market.
problem Impact of reference-dependent preferences on risk-return trade-off in Chinese stock market.
method Utilized CGO proxy, econometric techniques (Dependent Double Sorting, Fama-MacBeth regressions), and data from 1995-2024.
result Reference-dependent preferences have a weaker or absent positive risk-return relationship in the Chinese market.
Study shows competition feedback can make ML predictors biased towards specific user groups.
problem How competition affects machine learning predictors and user prediction quality.
method Flexible model of competing ML predictors, empirical and mathematical analysis.
result Competition causes predictors to specialize for specific sub-populations at the cost of general performance.
LLMs prefer Bitcoin under crisis frames, affecting financial decisions.
problem Testing whether LLMs have built-in biases towards specific financial assets.
method Developed a three-level audit protocol to examine Bitcoin's representation and influence in LLMs.
result An identifiable internal feature in LLMs can be perturbed to move financial choices, but only within measurable limits.
The paper presents a new method to represent directed graphs using pseudo-Riemannian manifolds.
problem Representing directed graphs in a compact and meaningful way.
method Combines pseudo-Riemannian metric structure, non-trivial global topology, and a unique likelihood function.
result Low-dimensional cylindrical Minkowski and anti-de Sitter spacetimes produce equal or better graph representations than curved Riemannian manifolds.
Differentiable NAS frameworks grow networks wider and deeper, revealing biases in wiring evolution.
problem Understanding the evolution of neural architecture wiring in differentiable NAS methods.
method Unified view on searching algorithms, local cost minimization, empirical and theoretical analyses.
result Implicit inductive biases cause observed searching patterns in differentiable NAS methods.
This paper evaluates debiasing methods on word embeddings to reduce religious bias.
problem Social biases persist in word embeddings, potentially amplifying them in AI applications.
method Investigates and evaluates three multiclass debiasing techniques on three word embeddings.
result ConceptorDebiasing is the most effective method, reducing religious bias by 82-96%.
We provide a mathematical definition of fragility and antifragility as negative or positive sensitivity to a semi-measure of dispersion and volatility (a variant of negative or positive "vega") and examine the link to nonlinear effects. We integrate model error (and biases) into the fragile or antifragile context. Unli…
This paper develops a model of reference-dependent assessment of subjective beliefs in which loss-averse people optimally choose the expectation as the reference point to balance the current felicity from the optimistic anticipation and the future disappointment from the realisation. The choice of over-optimism or over…
The first step in constructing a machine learning model is defining the features of the data set that can be used for optimal learning. In this work we discuss feature selection methods, which can be used to build better models, as well as achieve model interpretability. We applied these methods in the context of stres…
The paper shows how curated synthetic data can optimize human preferences in generative models.
problem Contamination of web-scale datasets by synthetic data affects future model training.
method Theoretical study of iterated retraining of generative models with curated synthetic data.
result Data curation can be seen as an implicit preference optimization mechanism, maximizing expected reward.
Paper proposes a data augmentation method for LLM-generated data in market research.
problem Bias in LLM-generated data in market research.
method Statistical data augmentation approach integrating LLM-generated and real data.
result Statistically robust estimators with reduced bias and cost savings.
We quantify content availability and user discovery opportunities in recommender systems.
problem Determining the maximum probability of recommending content to users.
method Stochastic reachability to compute upper bounds on recommendation likelihood.
result Reachability metrics can detect biases and diagnose user discovery limitations.
New algorithm ensures fairness without sacrificing accuracy.
problem Ensuring fairness in machine learning without harming accuracy.
method Demographic-Agnostic Fairness without Harm (DAFH) algorithm.
result DAFH algorithm achieves higher accuracy than existing methods.
Deep learning's anomalous generalization explained by standard frameworks.
problem Anomalous generalization in deep neural networks.
method Intuitive understanding and rigorous characterization using PAC-Bayes and countable hypothesis bounds.
result Deep learning's anomalous generalization can be explained by soft inductive biases.
The dynamics of many socioeconomic systems is determined by the decision making process of agents. The decision process depends on agent's characteristics, such as preferences, risk aversion, behavioral biases, etc.. In addition, in some systems the size of agents can be highly heterogeneous leading to very different i…
In this paper, we consider asymptotic properties of the support vector machine (SVM) in high-dimension, low-sample-size (HDLSS) settings. We show that the hard-margin linear SVM holds a consistency property in which misclassification rates tend to zero as the dimension goes to infinity under certain severe conditions. …
We generalise the problem of inverse reinforcement learning to multiple tasks, from multiple demonstrations. Each one may represent one expert trying to solve a different task, or as different experts trying to solve the same task. Our main contribution is to formalise the problem as statistical preference elicitation,…
Societies often rely on human experts to take a wide variety of decisions affecting their members, from jail-or-release decisions taken by judges and stop-and-frisk decisions taken by police officers to accept-or-reject decisions taken by academics. In this context, each decision is taken by an expert who is typically …
Drawing an inspiration from behavioral studies of human decision making, we propose here a general parametric framework for a reinforcement learning problem, which extends the standard Q-learning approach to incorporate a two-stream framework of reward processing with biases biologically associated with several neurolo…