This study finds ESG rating disagreement reduces corporate productivity, especially in certain types of firms.
problem The impact of ESG rating disagreement on corporate productivity.
method Analysis of A-share listed companies data from 2015 to 2022 using XGBoost regression and SHAP.
result ESG rating disagreement reduces corporate productivity, especially in certain types of firms.
Paper tackles ESG rating disagreement in sustainable investing portfolios.
problem Lack of alignment between ESG ratings from different agencies affects investment decisions.
method Proposes a nonlinear optimization model reformulated as a convex quadratic program to address ESG rating disagreement.
result The proposed model can effectively manage ESG rating disagreement and improve investment decisions.
Ensembling improves performance when classifiers disagree more than average.
problem When do ensembles provide significant performance improvements in classification tasks?
method Theoretical and empirical analysis of ensemble improvement rate and disagreement-error ratio.
result Ensembling improves performance significantly when the disagreement rate is large relative to the average error rate.
SGD-trained models' disagreement predicts test error.
problem Estimating test error of deep networks.
method Empirical testing and theoretical analysis of SGD ensembles.
result SGD ensembles' disagreement correlates with test error.
Vote-boosting uses weighted training data to build accurate and robust ensembles.
problem Generating accurate and robust ensemble classifiers.
method Sequential ensemble learning with weighted training data and emphasis on instances with high disagreement.
result Vote-boosting is effective for generating accurate and robust ensembles, especially when noise levels are low.
We study agnostic active learning, where the goal is to learn a classifier in a pre-specified hypothesis class interactively with as few label queries as possible, while making no assumptions on the true function generating the labels. The main algorithms for this problem are {\em{disagreement-based active learning}}, …
Theoretical analysis of co-training and disagreement-based algorithms.
problem Addressing theoretical issues in disagreement-based learning approaches.
method Theoretical analyses of co-training and its variants.
result Provides a theoretical foundation for co-training and similar algorithms.
Bayes-consistent disagreement discrepancy loss improves model robustness.
problem Distribution shift in real-world neural network deployment.
method Introducing a novel disagreement loss that is Bayes consistent.
result Proves existing surrogates for disagreement discrepancy are not Bayes consistent.
This paper characterizes and explains the disagreement between two graph embedding methods.
problem Understanding why two popular graph embedding methods produce different results.
method End-to-end analysis of ASE-LSE latent subspaces, proving conditions for agreement and disagreement.
result No maximal-disagreement graph exists; disagreement is strictly below its theoretical ceiling.
Financial economics provides intuition for the Rényi divergence.
problem Lack of practical intuition for statistical concepts like Rényi divergence.
method Using financial economics to transform disagreements into investment opportunities.
result The Rényi divergence's practical performance quantifies disagreement.
The study reveals a linear relationship between source and target domain classification errors based on disagreement.
problem Evaluating model performance under distribution shift with limited labeled data.
method Developed a theoretical foundation for analyzing disagreement in high-dimensional random features regression.
result The disagreement-on-the-line phenomenon occurs when classification error under the source domain is a linear function of the target domain.
Traditional multi-view learning approaches suffer in the presence of view disagreement,i.e., when samples in each view do not belong to the same class due to view corruption, occlusion or other noise processes. In this paper we present a multi-view learning approach that uses a conditional entropy criterion to detect v…
Improves reliability of medical diagnosis uncertainty estimates.
problem Label uncertainty in medical diagnosis.
method Post-hoc alpha-calibration method for neural network classifiers. result Significantly enhances reliability of uncertainty estimates.
Study investor sentiment and disagreement on StockTwits during COVID-19.
problem Understanding investor beliefs and sentiment during the pandemic.
method Analysis of social media data (StockTwits) for investor messages.
result Sentiment and disagreement sharply decreased in early March 2020, followed by a reversal.
This paper proposes a self-supervised exploration method using disagreement of dynamics models.
problem Efficient exploration in stochastic environments with real robots.
method Train an ensemble of dynamics models and incentivize exploration to maximize disagreement.
result Sample-efficient exploration achieved without external rewards or reinforcement learning.
Mitigates spurious correlations without bias labels.
problem Spurious correlations bias model performance.
method Introduces a novel training objective and debiasing method DPR.
result DPR achieves state-of-the-art performance.
Paper improves imitation learning from observations by minimizing inverse dynamics disagreement.
problem Challenges in Learning from Observations (LfO) due to incomplete expert guidance.
method IDDM method that minimizes the disagreement between inverse dynamics models of the imitator and expert.
result Empirical results show consistent improvements over other LfO methods.
This work optimizes clustering with adaptive queries to minimize disagreements.
problem Minimizing disagreements in clustering with adaptive similarity queries.
method Active learning algorithms and information-theoretical bounds.
result Achieves an almost optimal trade-off between queries and clustering error.
Learning with noisy labels is one of the hottest problems in weakly-supervised learning. Based on memorization effects of deep neural networks, training on small-loss instances becomes very promising for handling noisy labels. This fosters the state-of-the-art approach "Co-teaching" that cross-trains two deep neural ne…
This paper models how funds choose between competing ESG rating methodologies based on investor preferences.
problem Competing ESG rating methodologies lead to different portfolio rewards and fund fees.
method Modeling funds with heterogeneous ESG priorities and analyzing portfolio changes and investor demand.
result Funds specialize more, but provider scores, investor participation, and equilibrium fees decrease in the benchmark equilibrium.
New algorithms for learning under s-concave distributions, including Pareto and t-distributions.
problem Learning under broad and natural generalizations of log-concave distributions, including fat-tailed ones.
method Introduce new convex geometry tools to study s-concave distributions and use these properties to provide bounds on learning quantities. result Significantly generalize prior results for margin-based, disagreement-based, and passive learning of intersections of halfspaces.
We consider the problem of Probably Approximate Correct (PAC) learning of a binary classifier from noisy labeled examples acquired from multiple annotators (each characterized by a respective classification noise rate). First, we consider the complete information scenario, where the learner knows the noise rates of all…
This paper improves active learning by using robust divergences for committee disagreement.
problem Active learning with high measurement costs.
method Query by committee with Bregman divergence (including Kullback-Leibler divergence as a special case).
result The proposed method is more robust and performs as well as or better than conventional methods.
New algorithm improves active learning in agnostic pool-based classification.
problem Efficient active learning in the agnostic setting with minimized sample complexity.
method Solves an experimental design problem to determine a distribution over examples for label requests.
result Achieves sample complexity bounds never worse than best disagreement coefficient-based bounds, sometimes significantly smaller.
Structured credal learning separates covariate shift and label disagreement.
problem Uncertainty in real-world learning tasks due to covariate shift and noisy labels.
method Introduces a structured credal learning framework that explicitly separates these sources.
result Geometric bounds and decomposition reveal how covariate shifts affect label disagreement contributions.
Our work investigates disagreement in neural network feature attribution methods.
problem Disagreement among feature attribution methods for neural networks.
method Investigates the fundamental and distributional behavior of feature attribution methods.
result Illustrates the impact of scaling and encoding techniques on explanation quality.
The paper analyzes how multiple classifiers' disagreement and polarization affect overall accuracy.
problem Improving accuracy through ensembling multiple classifiers.
method The paper derives an upper bound for polarization, proposes a neural polarization law, and presents a tight upper bound for the error of majority vote classifiers.
result Disagreement and polarization among classifiers are linearly correlated with the target, and polarization is nearly constant for a dataset.
Two validation methods compared for classifier accuracy.
problem Validating trained classifiers using all in-sample data.
method SVOOSH vs. WAG (withhold and gap)
result WAG can be a better alternative for complex hypothesis classes and limited training data.
GRANITE unifies feature-based explanation methods to reduce disagreement.
problem Disagreement among feature-based explanation methods.
method GRANITE partitions feature space into regions minimizing interaction and distribution influences.
result Unified and consistent feature explanations.
Study compares and contrasts various ML explanation methods, highlighting their disagreements and similarities.
problem Understanding and quantifying the differences between various machine learning explanation methods.
method Synthesized and visualized various explanation methods for global and local aspects of ML models.
result There is substantial agreement on the top features but less on specific rankings, and tree interpreter is comparable to SHAP for feature effects.
EXAGREE selects a stakeholder-aligned model to reduce conflicting explanations in machine learning.
problem Conflicting explanations from different attribution methods limit the adoption of machine learning models in safety-critical domains.
method EXAGREE is a two-stage framework that selects a Stakeholder-Aligned Explanation Model (SAEM) from a set of similar-performing models, maximizing Stakeholder-Machine Agreement (SMA).
result EXAGREE achieves simultaneous gains in faithfulness, plausibility, and fairness over baselines while preserving task accuracy.
Models predict uncertainty to help identify patients needing second opinions.
problem Disagreements among medical experts, especially in diagnosis.
method Direct Uncertainty Prediction (DUP) training to predict uncertainty scores directly from patient features.
result DUP outperforms Uncertainty Via Classification in identifying cases needing second opinions.
AutoDIME automates design of multi-agent environments for RL.
problem Designing multi-agent environments for reinforcement learning is challenging.
method Developed intrinsic teacher rewards for multi-agent settings and evaluated them in various tasks.
result Value disagreement was found to be most consistent and effective across tasks.
New method gives provable error bounds for neural nets under distribution shift.
problem Proving reliable error bounds for neural networks under distribution shift.
method Optimizing a classifier to disagree with another, using a new 'disagreement loss'.
result Valid error bounds with comparable accuracy to competitive methods.
This work proves the optimal estimation rates for popular kernel discrepancies.
problem Estimating the disagreement of distributions using kernel discrepancies.
method Proving minimax lower bounds for MMD, HSIC, and KSD.
result The minimax lower bound for estimation of MMD, HSIC, and KSD is \( n^{-1/2} \) on general topological spaces.
New method reduces version space for CNNs, improving active learning performance.
problem Sampling bias in active learning hinders optimal hypothesis finding in neural networks.
method Version space reduction through prior mass reduction and diameter reduction, proposing a new Gibbs-vote disagreement method.
result Diameter-based querying method reduces version space more effectively than prior mass reduction and other methods.
New method pools labels from similar data items to improve learning from small samples.
problem Learning from small, human-annotated samples with potential disagreement among annotators.
method Proposes neighborhood-based pooling for sharing labels across similar data items.
result Improves learning from small, noisy samples by pooling labels from similar items.
The paper tackles fair correlation clustering with fairness constraints.
problem Minimizing disagreements while adhering to fairness constraints for clustering.
method Two variants of fairness constraints are considered: equal distribution and relative bounds. Approximation algorithms are developed for these constraints.
result Approximation algorithms for fair correlation clustering with theoretical guarantees and empirical validation.
Improves domain adaptation bounds using novel distribution pseudodistance.
problem Learning from source to target distributions with divergence controls.
method Introduces new bounds and learning algorithms based on disagreement averaging.
result PAC-Bayesian bounds for target risk with distribution divergence controls.
CreDRO learns credal ensembles via distributionally robust optimization, improving EU quantification.
problem Quantifying predictive epistemic uncertainty in credal models.
method Distributionally robust optimization to capture EU from training randomness and potential distribution shifts.
result Empirically, CreDRO outperforms existing credal methods on various tasks.
The paper explores how Shapley value for a feature can vary based on model outcomes and feature distribution.
problem The uniqueness of Shapley value in explaining model predictions.
method Analyzes the relationship between feature distribution and Shapley value, and compares Shapley values for different model outcomes.
result Shapley value for a feature depends on more than just its mean and can vary significantly based on model outcome.
Algorithm identifies controversial regions in classifier disagreements.
problem Understanding regions of high disagreement among classifiers.
method Exceptional Model Mining framework.
result Shows usefulness in identifying new phenomena in well-explored datasets.
We introduce a new and improved characterization of the label complexity of disagreement-based active learning, in which the leading quantity is the version space compression set size. This quantity is defined as the size of the smallest subset of the training data that induces the same version space. We show various a…
We show that the disagreement coefficient of certain smooth hypothesis classes is O(m), where m is the dimension of the hypothesis space, thereby answering a question posed in \cite{friedman09}.
This paper relaxes the common prior assumption in the public and private information game of Morris and Shin (2000, 2004). For the generalized game, where the agent's prior expectations are heterogenous, it derives a sharp condition for the emergence of unique/multiple equilibria. This condition indicates that unique e…
Paper presents a defense framework against adversarial examples.
problem Vulnerability of deep neural networks to adversarial examples.
method Cross-layer strategic ensemble defense with input and output transformations.
result Strategic ensemble defense achieves high defense success rates and robustness.
Online monitor detects classifier drift and adapts predictions.
problem Silent degradation of classifier accuracy under distributional shift.
method Sliding-window KS statistic with calibrated alarm thresholds.
result 86.6% valid detection across various shift conditions.
New method boosts skill learning by encouraging optimistic exploration.
problem Intrinsic reward for exploration is inherently pessimistic.
method Derive an information gain auxiliary objective involving an ensemble of discriminators and rewarding policy disagreement.
result Improves skill learning in grid worlds and Atari games.