Study of loss functions for learning to defer, proving consistency.
problem Learning to defer in machine learning.
method Introduced a family of surrogate losses parameterized by Ψ and proved their consistency. result Proved realizable H-consistency and Bayes-consistency of specific surrogate losses. In this paper we study the consistency of an empirical minimum error entropy (MEE) algorithm in a regression setting. We introduce two types of consistency. The error entropy consistency, which requires the error entropy of the learned function to approximate the minimum error entropy, is shown to be always true if the…
This paper improves deep learning model consistency through ensemble methods.
problem Consistency and correct-consistency issues in deep learning models.
method Formal definition of consistency and correct-consistency, proving ensemble improvement, proposing dynamic snapshot ensemble method.
result Ensemble methods can improve correct-consistency of deep learning models.
Empirical study shows consistent meta-RL algorithms adapt to OOD tasks.
problem Theoretical consistency of meta-RL algorithms and its practical implications.
method Empirical investigation of representative meta-RL algorithms, focusing on consistency and adaptation to out-of-distribution tasks.
result Theoretical consistent algorithms can adapt to OOD tasks, while inconsistent ones cannot, but can still fail for poor exploration.
Paper explores grafting consistent estimators to improve Random Forest consistency.
problem Ensuring Random Forests are consistent despite their performance.
method Grafting consistent estimators onto a shallow CART.
result Grafted estimators provide a consistency guarantee and perform well empirically.
Unified model improves sampling speed and quality.
problem Difficult to balance sampling speed and quality.
method Multistep Consistency Models combining consistency and diffusion models.
result Improved sampling quality with reduced steps.
Paper connects risk consistency to L_p consistency for broader loss functions.
problem Establishing risk consistency for a wider class of loss functions.
method Analyzes the connection between risk consistency and L_p-consistency for various loss functions.
result Shifted loss functions do not reduce assumptions as much as other results.
Improves GAN-based semi-supervised learning with consistency regularization.
problem Lack of consistency in class probability predictions under local perturbations.
method Introduces consistency regularization to GANs, leveraging both local and interpolation consistency.
result Significantly improves performance and achieves new state-of-the-art results.
This paper establishes a theoretical foundation for consistency training in diffusion models.
problem Lack of a comprehensive theoretical understanding of consistency training in diffusion models.
method Demonstrates the necessity of a number of steps in consistency learning exceeding d5/2/ε for generating samples within ε proximity to the target distribution. result Establishes rigorous insights into the validity and efficacy of consistency models, offering theoretical underpinnings for their utility.
Paper introduces new actuarial-consistent valuations for insurance liabilities.
problem Valuation of insurance liabilities considering both financial and actuarial risks.
method Proposes two-step actuarial valuations and actuarial-consistent procedures.
result Actuarial-consistent valuations are equivalent to two-step actuarial valuations under coherence.
This paper aims to analyze knowledge consistency between pre-trained deep neural networks. We propose a generic definition for knowledge consistency between neural networks at different fuzziness levels. A task-agnostic method is designed to disentangle feature components, which represent the consistent knowledge, from…
Kleinberg introduced three natural clustering properties, or axioms, and showed they cannot be simultaneously satisfied by any clustering algorithm. We present a new clustering property, Monotonic Consistency, which avoids the well-known problematic behaviour of Kleinberg's Consistency axiom, and the impossibility resu…
We improve random forest consistency and performance with DMRF, a new variant.
problem Improving the consistency and performance of random forest models.
method Developed DMRF, a data-driven multinomial random forest, by modifying proof methods and improving data utilization.
result DMRF achieves strong consistency with probability 1, surpassing previous models in classification tasks.
Study of hill-climbing clustering methods and their consistency.
problem Consistency of hill-climbing clustering methods.
method Continuous-space and discrete-space hill-climbing approaches.
result Established consistency of the methods.
This paper tackles deferral learning with multiple experts, providing strong theoretical guarantees.
problem Optimizing input assignment to experts balancing accuracy and computational cost.
method Introducing new surrogate loss functions and efficient algorithms with strong theoretical learning guarantees.
result Realizable H-consistency, H-consistency bounds, and Bayes-consistency for deferral learning. Unified surrogate loss framework for multi-label learning with strong consistency guarantees.
problem Improving consistency and accounting for label correlations in multi-label learning.
method Introducing multi-label logistic loss and extending it to comprehensive multi-label comp-sum losses, proving strong consistency guarantees for any multi-label loss.
result Unified surrogate loss framework benefiting from strong consistency guarantees for any multi-label loss.
Learning rule consistency tied to non-existence of real-valued measurable cardinals.
problem Consistency of k-NN learning rule in metric spaces.
method Analyzing separable subspaces and density conditions.
result The k-NN classifier's consistency depends on the absence of real-valued measurable cardinals.
New method evaluates language model forecasters by checking consistency of predictions.
problem Evaluating the performance of language model forecasters is difficult due to lack of ground truth.
method Developed a consistency check framework based on arbitrage to evaluate forecasters.
result Consistency metrics correlate with ground truth performance of LLM forecasters.
Study improves H-consistency bounds for regression analysis.
problem Improving H-consistency bounds for regression analysis. method Generalized theorems and novel H-consistency bounds for various surrogate loss functions. result Derives principled surrogate losses for adversarial regression.
MTSCI uses diffusion models to impute multivariate time series data with consistency.
problem Imputation of missing values in multivariate time series data.
method MTSCI employs a contrastive complementary mask and mixup mechanism to ensure intra-consistency and inter-consistency.
result MTSCI achieves state-of-the-art performance on multivariate time series imputation tasks.
Self-consistent models improve reinforcement learning by aligning predictions with future values.
problem Improving reinforcement learning by aligning model predictions with future values.
method Proposes multiple self-consistency updates to encourage a learned model and value function to be consistent with each other.
result Self-consistency helps both policy evaluation and control in both tabular and function approximation settings.
Prefix consistency improves model reliability by weighting answers based on their reproducibility.
problem Improving the reliability of large language models' reasoning traces.
method Use prefix consistency to weight candidate answers based on their reproducibility during regeneration.
result Prefix consistency is the best correctness predictor, reaching Standard MV plateau accuracy with up to 21x fewer tokens.
The paper concerns primal and dual representations as well as time consistency of set-valued dynamic risk measures. Set-valued risk measures appear naturally when markets with transaction costs are considered and capital requirements can be made in a basket of currencies or assets. Time consistency of scalar risk measu…
This paper establishes the asymptotic consistency of the {\it loss-calibrated variational Bayes} (LCVB) method. LCVB was proposed in~\cite{LaSiGh2011} as a method for approximately computing Bayesian posteriors in a `loss aware' manner. This methodology is also highly relevant in general data-driven decision-making con…
The paper establishes conditions for Bayesian consistency in supremum metric.
problem Ensuring Bayesian consistency in the supremum metric.
method Using a triangle inequality and weak convergence, the paper establishes conditions for Bayesian consistency.
result Demonstrates supremum consistency with weaker conditions than previously used.
Consistency regularization improves robustness to noisy labels.
problem Improving model robustness to noisy labels in machine learning.
method Empirical study of consistency regularization on noisy datasets.
result Consistency regularization improves model robustness to label noise.
Enhances random forest consistency and introduces DMRF for improved performance.
problem Improving the consistency and efficiency of random forest algorithms.
method Strengthened proof methods and propose DMRF algorithm.
result DMRF achieves better theoretical and experimental performance than previous variants.
Adversarial consistency depends on the uniqueness of adversarial Bayes classifiers.
problem Consistency of adversarial surrogate losses is not guaranteed.
method Connected consistency of adversarial surrogate losses to the uniqueness of adversarial Bayes classifiers.
result A convex surrogate loss is statistically consistent for adversarial learning if and only if the adversarial Bayes classifier is unique.
Develops PLL methods that are provably consistent and compatible with any deep network.
problem Lack of theoretical understanding and consistency in partial-label learning methods.
method Proposes a generation model for candidate label sets and develops two PLL methods that are risk- and classifier-consistent.
result Two novel PLL methods are guaranteed to be provably consistent.
We extend a recently proposed 1-nearest-neighbor based multiclass learning algorithm and prove that our modification is universally strongly Bayes-consistent in all metric spaces admitting any such learner, making it an "optimistically universal" Bayes-consistent learner. This is the first learning algorithm known to e…
New method improves consistency in preference learning for neural networks.
problem Inconsistent surrogate losses in preference learning for neural networks.
method Formulated a margin-shifted ranking framework and introduced Structure-Aware H-consistency. result Proved superior consistency guarantees for capacity-bounded models using heavy-tailed surrogates.
ε-Consistent Mixup improves semi-supervised classification accuracy.
problem Improving semi-supervised classification accuracy with limited labeled data.
method Combines Mixup's linear interpolation with consistency regularization, using an adaptive tradeoff between the two.
result ε-Consistent Mixup yields the largest gains in low label-availability scenarios.
Study shows k-NN regressor consistency in complex survey designs.
problem Lack of consistency results for k-NN regressor in complex survey data. method Analysis of regularity conditions on sampling design and data distribution.
result Consistency of k-NN regressor under complex survey designs. Proves DCNNs with expansive convolution are strongly universally consistent.
problem Theoretical consistency of deep convolutional neural networks (DCNNs).
method Empirical risk minimization on DCNNs with expansive convolution (with zero-padding).
result DCNNs with expansive convolution are strongly universally consistent.
Regularized M-estimators are used in diverse areas of science and engineering to fit high-dimensional models with some low-dimensional structure. Usually the low-dimensional structure is encoded by the presence of the (unknown) parameters in some low-dimensional model subspace. In such settings, it is desirable for est…
We consider families of strongly consistent multivariate conditional risk measures. We show that under strong consistency these families admit a decomposition into a conditional aggregation function and a univariate conditional risk measure as introduced Hoffmann et al. (2016). Further, in analogy to the univariate cas…
MaxMatch improves SSL with worst-case consistency for better generalization.
problem Efficiently supervised learning with unlabeled data.
method Worst-case consistency regularization for SSL, providing a bound and an algorithm.
result The proposed method converges to a stationary point and improves generalization.
The paper examines the consistency of Lasso regression applied to signature analysis of time series data.
problem Consistency of Lasso regression in signature analysis of time series data.
method The paper studies the consistency of Lasso regression applied to signature analysis of time series data, both theoretically and numerically.
result The Lasso regression is consistent both asymptotically and in finite sample for certain types of time series and processes.
In this paper we provide a flexible framework allowing for a unified study of time consistency of risk measures and performance measures (also known as acceptability indices). The proposed framework not only integrates existing forms of time consistency, but also provides a comprehensive toolbox for analysis and synthe…
Study on H-consistency bounds for machine learning surrogates.
problem Estimating target loss error relative to surrogate loss error in machine learning.
method Developed H-consistency bounds for various surrogates and loss functions. result Stronger guarantees than existing methods, offering distribution-dependent and -independent bounds.
In this paper we present results on dynamic multivariate scalar risk measures, which arise in markets with transaction costs and systemic risk. Dual representations of such risk measures are presented. These are then used to obtain the main results of this paper on time consistency; namely, an equivalent recursive form…
New rule universally consistent for online learning with non-ergodic data.
problem Online learning with non-ergodic data processes.
method Developed an online learning rule for processes on (X,Y) pairs.
result Generalizes past results to non-ergodic processes on (X,Y).
We consider evaluation methods for payoffs with an inherent financial risk as encountered for instance for portfolios held by pension funds and insurance companies. Pricing such payoffs in a way consistent to market prices typically involves combining actuarial techniques with methods from mathematical finance. We prop…
Study assesses consistency and reproducibility of LLMs in finance and accounting tasks.
problem Consistency and reproducibility of LLM outputs in finance and accounting research.
method Extensive experimentation with 50 independent runs across 5 tasks using 3 OpenAI models.
result Task-specific patterns of consistency and reproducibility, with binary classification and sentiment analysis achieving near-perfect reproducibility.
MIC consistently estimates dependence in large datasets.
problem Estimating dependence between variable pairs in large datasets.
method Proving consistency of MIC as an estimator.
result MIC is a consistent estimator of population statistic MIC*.
In this paper we present results on scalar risk measures in markets with transaction costs. Such risk measures are defined as the minimal capital requirements in the cash asset. First, some results are provided on the dual representation of such risk measures, with particular emphasis given on the space of dual variabl…
The problem of clustering is considered, for the case when each data point is a sample generated by a stationary ergodic process. We propose a very natural asymptotic notion of consistency, and show that simple consistent algorithms exist, under most general non-parametric assumptions. The notion of consistency is as f…
The problem of clustering is considered, for the case when each data point is a sample generated by a stationary ergodic process. We propose a very natural asymptotic notion of consistency, and show that simple consistent algorithms exist, under most general non-parametric assumptions. The notion of consistency is as f…