arXiv research

Many real-world analytics problems involve two significant challenges: prediction and optimization. Due to the typically complex nature of each challenge, the standard paradigm is predict-then-optimize. By and large, machine learning tools are intended to minimize prediction error and do not account for how the predict…

This paper improves risk bounds and calibration for smart predict-then-optimize method.

Develops active learning method for linear optimization with margin-based criterion.

New loss function handles uncertain constraints in CSLO problems.

The predict-then-optimize framework is fundamental in many practical settings: predict the unknown parameters of an optimization problem, and then solve the problem using the predicted values of the parameters. A natural loss function in this environment is to consider the cost of the decisions induced by the predicted…

A new autoregressive SPO method improves decision-making for dependent data.

Decision trees improve decision-making by optimizing predictions of unknown parameters.

Paper proposes SPO paradigm for better portfolio optimization in real markets.

This study explains and mitigates inflated returns and turnover in SPO-based portfolio optimization.

SPO optimizes LLMs by eliminating group-based baselines and variance issues.

Stochastic particle-optimization sampling (SPOS) is a recently-developed scalable Bayesian sampling framework that unifies stochastic gradient MCMC (SG-MCMC) and Stein variational gradient descent (SVGD) algorithms based on Wasserstein gradient flows. With a rigorous non-asymptotic convergence theory developed recently…

We consider the space of differential operators $\mathcal{D}_{λμ}$ acting between $λ$- and $μ$-densities defined on $S^{1|2}$ endowed with its standard contact structure. This contact structure allows one to define a filtration on $\mathcal{D}_{λμ}$ which is finer than the classical one, obtained by writting a differen…

Proposes BU-SPO method to improve text classification robustness.

The paper optimizes portfolios by selecting financial ratios via PCA for better value investment.

In this paper, we show that the Lie superalgebra $\mathfrak{spo}(2l+2|n)$ is into the intersection of Lie superalgebra of contact vector fields $\mathcal{K}(2l+1|n)$ and the Lie superalgebra of projective vector fields $\mathfrak{pgl}(2l+2|n)$. We use mainly the embedding used by P. Mathonet and F. Radoux in "\textit{ …

dboost optimizes prediction models for convex cone problems.

A smart method predicts and optimizes decisions online with resource constraints.

We analyse perception and memory, using mathematical models for knowledge graphs and tensors, to gain insights into the corresponding functionalities of the human mind. Our discussion is based on the concept of propositional sentences consisting of \textit{subject-predicate-object} (SPO) triples for expressing elementa…

In this paper, we generalize the known results on the super circles $S^{1|1}$ and $S^{1|2}$. We construct the fine equivariant quantization on the super circle $S^{1|n}$ for $n\geqslant 3$. The equivariant Lie superalgebra is $\spo(2|n)$ which is constituted of the contact projective vector fields on $S^{1|n}$. In orde…

Particle-optimization-based sampling (POS) is a recently developed effective sampling technique that interactively updates a set of particles. A representative algorithm is the Stein variational gradient descent (SVGD). We prove, under certain conditions, SVGD experiences a theoretical pitfall, {\it i.e.}, particles te…

New method identifies latent treatment effects from proxy models.

DFFL tackles federated learning with heterogeneous objectives and constraints.

Improves drug properties using a novel LLM and reinforcement learning.

RL models outperform traditional methods in certain market conditions.

New method optimizes policies without assuming known link functions between preferences and rewards.

The goal of machine learning is to provide solutions which are trained by data or by experience coming from the environment. Many training algorithms exist and some brilliant successes were achieved. But even in structured environments for machine learning (e.g. data mining or board games), most applications beyond the…

We consider the standard contact structure on the supercircle, S^{1|1}, and the supergroups E(1|1), Aff(1|1) and SpO(2|1) of contactomorphisms, defining the Euclidean, affine and projective geometry respectively. Using the new notion of (p|q)-transitivity, we construct in synthetic fashion even and odd invariants chara…

We study in this paper, the existence of tree types of filtrations of the space $\mathcal{D}_{λμ}(\R^{2l+1|n})$ of differential operators on the superspaces $\R^{2l+1|n}$ endowed with the standard contact structure $α$. On this space $\mathcal{D}_{λμ}(\R^{2l+1|n})$, we have the first filtration called canonical and bec…

Introduces Fitzpatrick losses, tighter than Fenchel-Young losses.

We study losses for binary classification and class probability estimation and extend the understanding of them from margin losses to general composite losses which are the composition of a proper loss with a link function. We characterise when margin losses can be proper composite losses, explicitly show how to determ…

We present the Tamed Cross Entropy (TCE) loss function, a robust derivative of the standard Cross Entropy (CE) loss used in deep learning for classification tasks. However, unlike other robust losses, the TCE loss is designed to exhibit the same training properties than the CE loss in noiseless scenarios. Therefore, th…

Unified surrogate loss framework for multi-label learning with strong consistency guarantees.

This paper introduces new loss functions for balanced multi-class classification.

We present $α$-loss, $α\in [1,\infty]$, a tunable loss function for binary classification that bridges log-loss ($α=1$) and $0$-$1$ loss ($α= \infty$). We prove that $α$-loss has an equivalent margin-based form and is classification-calibrated, two desirable properties for a good surrogate loss function for the ideal y…

This work broadens calibeating to various proper losses using Bregman divergence.

This work generalizes calibeating for a broader range of proper losses using Bregman divergence.

Proposes squentropy loss for improved classification accuracy and model calibration.

New loss function calibrates WW-hinge loss for multiclass SVM.

The study analyzes a model for aggregate losses with dependent and overdispersed inter-losses times.

Symmetric losses improve classifier robustness from corrupted labels.

Two new algorithms improve performance in adversarial bandits with unbounded losses.

Paper explores connections between loss functions and consistency in binary classification and regression.

Paper introduces a new topological loss for better convergence.

Novel loss functions improve decision tree learning from noisy data.

Theoretical analysis of cross-entropy loss functions and their robustness.

This paper improves operational risk modeling by selecting better loss severity distributions.

This paper improves loss functions for deep learning with noisy labels.

Classification is the most important process in data analysis. However, due to the inherent non-convex and non-smooth structure of the zero-one loss function of the classification model, various convex surrogate loss functions such as hinge loss, squared hinge loss, logistic loss, and exponential loss are introduced. T…