The Rashomon effect shows many models can perform similarly, explored in this paper.
problem Why do many models perform similarly in machine learning?
method Categorized causes into statistical, structural, and procedural sources.
result Structural multiplicity persists and cannot be resolved without additional assumptions.
Commentary on Rashomon Effect complicating model selection.
problem Many models equally predict data; hard to draw conclusions.
method Connections to recent ML literature exploring implications.
result Grasping Rashomon Effect can foster collaboration.
New algorithms approximate Rashomon set for sparse models, aiding expert interaction.
problem Lack of interaction between models and domain experts in classical machine learning.
method Approximate Rashomon set of sparse, generalized additive models using ellipsoids.
result Efficiently approximated Rashomon set facilitates model selection and exploration.
Framework uses dropout to efficiently explore Rashomon set for multiplicity estimation.
problem Efficiently measuring and mitigating conflicting model outputs in classification tasks.
method Dropout-based exploration of Rashomon set for multiplicity estimation.
result Framework outperforms baselines in multiplicity metric estimation with significant runtime speedup.
Study shows more data improves model explanations, aiding reliable knowledge extraction.
problem Challenges in deriving reliable knowledge from machine learning models due to the Rashōmon effect.
method Examined the influence of sample size on explanations from models in a Rashōmon set using SHAP.
result Explanations from <128 samples are highly variable, but agreement improves with more data.
New method uses Rashomon sets to improve Bayesian inference in factorial designs.
problem Combustion of model uncertainty in factorial designs leads to multimodal posterior and convergence issues.
method Rashomon-seeded annealing, integrating high-performing models as warm start for AIS.
result Restores full posterior inference without exhaustive enumeration of model space.
Paper explores Rashomon set models for more trustworthy medical conclusions.
problem Lack of comprehensive analysis of models in Rashomon set leads to misleading conclusions.
method Introduces Rashomon_DETECT algorithm and Profile Disparity Index (PDI).
result Combining differently behaving models in Rashomon set provides more trustworthy conclusions.
Leo Breiman's Rashomon Effect and Occam Dilemma are re-evaluated in the context of modern machine learning.
problem The tradeoff between model complexity and accuracy in machine learning.
method Modern perspective on Breiman's arguments using current computational capabilities.
result Algorithmic models can be accurate without being complex, nullifying the Occam Dilemma.
Noise increases the Rashomon ratio, leading simpler models to perform similarly to complex ones.
problem Why simpler models perform similarly to complex models on noisy datasets.
method Analyzed the data generation process and model training choices, introduced pattern diversity.
result Noisier datasets lead to larger Rashomon ratios, explaining simpler models' performance.
The paper analyzes the Rashomon ratio for infinite classifier families and shows its importance for choosing good classifiers.
problem Analyzing the Rashomon ratio for infinite classifier families.
method Quantifying the Rashomon ratio in two examples and providing guarantees for estimating it.
result A large Rashomon ratio guarantees choosing a classifier with good empirical accuracy will not significantly increase empirical loss.
Paper presents a Rashomon Quartet to show different model explanations.
problem Models with similar performance can explain data differently.
method Synthetic dataset with four equally effective models.
result Models with similar performance can explain data differently.
Introduces Rashomon Capacity to measure predictive multiplicity in probabilistic classifiers.
problem Predictive multiplicity in classification models leading to unjustified decisions.
method Introduces Rashomon Capacity, a metric for probabilistic classifiers, and provides a rigorous derivation.
result Rashomon Capacity captures nuanced score variations and provides strategies for disclosing conflicting models.
AXE evaluates explanations to avoid misleading Rashomon set model selection.
problem Evaluating explanations for Rashomon set models to avoid false selection.
method Proposed AXE method to evaluate explanation quality.
result AXE detects adversarial fairwashing with 100% success rate.
REALITrees uses a Rashomon ensemble approach for active learning in sparse decision trees.
problem Active learning reduces labeling costs by selecting informative samples, but current methods often sacrifice model diversity and direct characterization of the hypothesis space.
method REALITrees constructs a committee of all near-optimal sparse decision tree models using a Rashomon Set and a Gibbs posterior to weight them by empirical risk.
result REALITrees outperforms randomized ensembles, especially in noisy environments, by leveraging expanded model multiplicity.
It is almost always easier to find an accurate-but-complex model than an accurate-yet-simple model. Finding optimal, sparse, accurate models of various forms (linear models with integer coefficients, decision sets, rule lists, decision trees) is generally NP-hard. We often do not know whether the search for a simpler m…
This paper explores how balancing and filtering techniques affect predictive multiplicity in machine learning models.
problem Predictive multiplicity due to Rashomon effect in high-stakes environments.
method Investigates the impact of balancing and filtering techniques on predictive multiplicity using 21 real-world datasets.
result Data-centric AI strategies can mitigate predictive multiplicity, but preprocessing methods may introduce it.
Bayesian framework for model uncertainty identifies complex heterogeneity without strong assumptions.
problem Identifying complex heterogeneity in factorial data with varying covariates.
method Rashomon Partition Sets (RPS) using l0 prior for robust model uncertainty.
result RPS provides a robust set of models capturing complex heterogeneity without strong assumptions.
UNREAL selectively ensembles distinct models to improve active learning performance.
problem Difficulty in distinguishing genuine uncertainty from noise in limited labeled data.
method Selective ensembling of distinct models from the Rashomon set.
result UNREAL achieves faster convergence and up to 20% predictive accuracy improvement.
Framework for assessing fairness across similar predictive models.
problem Fairness in predictive models across different groups.
method Develops a framework for characterizing fairness over the set of good models under selective labels.
result Framework can replace or audit models for better fairness properties.
New framework quantifies variable importance across all good models and is stable across data distribution.
problem Conflicting variable importance conclusions from different models trained on the same data.
method Proposes a new variable importance framework that considers all good models and is stable across data distribution.
result Framework accurately estimates true variable importance and recovers rankings for complex setups.
Breiman's data analysis dichotomy is outdated, offering a third approach: mechanistic models.
problem Data analysis dichotomy between data modelers and algorithmic modelers.
method Interpolating between simple interpretable models and flexible function approximations using mechanistic models.
result Flexible, interpretable, and scientifically-informed hybrids can provide accurate and robust predictions.
Interactive EMA combines multiple explainability methods to improve model understanding.
problem Isolated explanations of machine learning models lead to misunderstandings and wrong reasoning.
method Interactive EMA (IEMA) combines multiple explainability methods sequentially.
result Interactive EMA increases the performance and confidence of human decision making.
This work introduces robust counterfactuals for neural networks that remain valid after minor model changes.
problem Generating counterfactual explanations that are robust to slight model changes.
method Introduces a measure called Stability to quantify robustness of counterfactuals to potential model changes, and proposes practical relaxations.
result Counterfactuals with high Stability remain valid after potential model changes with high probability.
A GPU framework speeds up BnB for discrete optimization problems.
problem Optimizing large-scale discrete problems with GPU limitations.
method Parallel BnB nodes in GPU batches, using padding and custom kernels.
result One to two orders of magnitude speedup and zero optimality gap.
Survey on principles and challenges of interpretable machine learning.
problem Improving machine learning models' interpretability for high-stakes decisions.
method Identification and analysis of 10 technical challenges in interpretable machine learning.
result Identification of 10 technical challenges in interpretable machine learning.
Random projections improve classifier generalization without needing to choose the best threshold.
problem Improving classifier generalization without choosing the best threshold.
method Thresholding a random one-dimensional feature after random projection of data.
result Generalization gap is significantly smaller than linear classifiers.
The leverage effect refers to the generally negative correlation between the return of an asset and the changes in its volatility. There is broad agreement in the literature that the effect should be present for theoretical reasons, and it has been consistently found in empirical work. However, a few papers have pointe…
New method for interpreting non-linear models using forward marginal effects.
problem Interpreting non-linear models' feature effects is challenging.
method Introducing forward marginal effects and partitioning feature space for better interpretation.
result Improved interpretation of non-linear prediction functions.
The Kalinin effectivity is studied and applied to compactifications and Hilbert squares.
problem Understanding Kalinin effectivity in compactifications and its applications.
method Definition, construction methods, and analysis of Kalinin effectivity in various compactifications.
result Wonderful compactifications of hyperplane arrangements and configuration spaces are Kalinin effective.
New method estimates treatment effects in network data, accounting for spillover effects.
problem Treatment effect estimation in networks with spillover effects.
method Augmented inverse probability weighting (AIPW) with cross-fitting and machine learning.
result Semiparametric treatment effect estimator converges at parametric rate and follows Gaussian distribution.
Causalfe estimates treatment effects in panel data with fixed effects.
problem Spurious heterogeneity in treatment effect estimates due to fixed effects in panel data.
method CFFE approach with node-level residualization during tree construction.
result Validates the estimator's performance through simulation studies.
The paper clarifies the distinction between CATE and ITE under ignorability assumptions.
problem Confusion between CATE and ITE hinders personalized effect estimation.
method Clarifies the distinction between CATE and ITE under ignorability assumptions.
result CATE and ITE are not necessarily the same under ignorability assumptions.
GADGET framework decomposes global feature effects using recursive partitioning.
problem Misleading global feature effects when feature interactions are present.
method Generalized additive decomposition of global effects (GADGET) based on recursive partitioning.
result Minimizes interaction-related heterogeneity of local feature effects.
A new RL framework evaluates dynamic mediation effects over time.
problem Dynamic mediation effects in sequentially assigned treatments.
method Reinforcement Learning framework for decomposition and estimation of causal effects.
result Superior performance demonstrated through numerical studies and real data analysis.
New memory effect discovered in gravitational wave behavior.
problem Understanding gravitational wave behavior in spacetimes with angular momentum.
method Mathematical analysis of Minkowski spacetime and Kerr black holes.
result Angular momentum memory effect observed at future null infinity.
Effective Yau-Tian-Donaldson conjecture for spherical varieties.
problem Finding effective K-stability criteria for spherical varieties.
method Formulated an effective variant of the Yau-Tian-Donaldson conjecture and reviewed effective K-stability criteria for spherical varieties.
result Effective K-stability criteria can be computed given combinatorial data.
A new method estimates treatment effects in mixed groups, improving accuracy.
problem Estimating treatment effects in mixed groups with heterogeneous responses.
method PCM (pre-cluster and merge) approach for nonparametric estimation.
result Asymptotic consistency and significant improvement in accuracy over existing methods.
Study estimates heterogeneous principal causal effects with binary treatments and intermediate variables.
problem Estimating subgroup effects within strata defined by potential values of an intermediate variable.
method Proposes a framework for estimating and forming confidence intervals for heterogeneous principal causal effects under principal ignorability assumption. Develops several estimators with varying robustness properties.
result Established large-sample theory and analyzed bias contributions of each approach.
Study develops method for estimating causal effects in continuous variables.
problem Lack of methods for estimating causal effects in continuous variables.
method Develops a method independent of data generating models for continuous variable interventions.
result Preserves identifiability of data and applies to any generating models.
ICA accurately estimates treatment effects even with confounders.
problem Estimating treatment effects in the presence of confounding variables.
method Uses Independent Component Analysis (ICA) to identify latent sources and estimate mixing coefficients.
result Linear ICA can consistently estimate multiple treatment effects, even with Gaussian confounders, and is more sample-efficient than Orthogonal Machine Learning (OML).
Method improves treatment effect estimation in randomized experiments.
problem Estimating distributional treatment effects in randomized experiments.
method Distributional regression framework with machine learning for variance reduction.
result The proposed method reduces variance of distributional treatment effect estimators.
New method discovers context effects in choice data.
problem Identifying context effects from choice data is challenging.
method Automatic discovery of context effects from observed choices.
result Automatic discovery of context effects from observed choices.
Dropout introduces both explicit and implicit regularization effects.
problem Understanding the full impact of dropout regularization.
method Disentangled explicit and implicit regularization effects through experiments and analytic simplifications.
result Explicit and implicit regularization effects of dropout are distinct and can be characterized analytically.
New test for binary treatment effects using kernel methods.
problem Testing distributional effects of binary treatments.
method Kernel-based doubly-robust test, avoiding permutations.
result Valid type-I error with computational efficiency.
New Krylov subspace methods speed up mixed-effects models with crossed random effects.
problem Slow computations for high-dimensional crossed random effects in mixed-effects models.
method Krylov subspace-based methods for generalized mixed-effects models with cross effects.
result Speedups by factors of up to 10,000 in computations for mixed-effects models.
The use of drug combinations, termed polypharmacy, is common to treat patients with complex diseases and co-existing conditions. However, a major consequence of polypharmacy is a much higher risk of adverse side effects for the patient. Polypharmacy side effects emerge because of drug-drug interactions, in which activi…
Previous literature has identified an effect, dubbed the Zumbach effect, that is nonzero empirically but conjectured to be zero in any conventional stochastic volatility model. Essentially this effect corresponds to the property that past squared returns forecast future volatilities better than past volatilities foreca…
Identifies causal effects in LiNGAM models with latent variables.
problem Identifying causal effects in LiNGAM models with latent confounders.
method Complete graphical characterization and efficient algorithms for certification. RICA adaptation for estimation.
result Efficient algorithms and RICA adaptation for estimating causal effects.