Study of measured laminations on surfaces using Newton polytopes and Poisson brackets.
problem Understanding the space of measured laminations on surfaces from a valuative perspective.
method Introducing Newton polytopes for character variety functions, defining tangent spaces, and identifying symplectic structures.
result Trace functions have unit coefficients at the extremal points of their Newton polytopes.
We are motivated by problems that arise in a number of applications such as Online Marketing and Explosives detection, where the observations are usually modeled using Poisson statistics. We model each observation as a Poisson random variable whose mean is a sparse linear superposition of known patterns. Unlike many co…
Developed shrinkage methods for Poisson regression models with experts to handle multicollinearity.
problem Multicollinearity in Poisson regression models with experts.
method Ridge and Liu-type shrinkage methods.
result Shrinkage methods offer more reliable estimates for coefficients in multicollinearity.
Our work is focused on the joint sparsity recovery problem where the common sparsity pattern is corrupted by Poisson noise. We formulate the confidence-constrained optimization problem in both least squares (LS) and maximum likelihood (ML) frameworks and study the conditions for perfect reconstruction of the original r…
Test evaluates NMF-based topic models for document corpora.
problem Violation of likelihood assumptions in NMF topic models.
method Double parametric bootstrap test based on KL divergence and Poisson ML.
result Correctly identifies reliable NMF-based topic models.
Proposes ML methods for robust price-sensitivity estimation in dynamic pricing.
problem Estimating price elasticities robustly in the presence of feature-dependent sensitivity.
method Poisson semi-parametric model with two-stage estimation: first-stage ML for observed purchases, second-stage Bayesian GLM for price-sensitivity.
result Reduces estimation error in price-sensitivity parameters from 25% to 4%.
BlinkML speeds up ML training by 6x-600x with probabilistic guarantees.
problem Ad-hoc sampling in ML analysis leads to unreliable model quality.
method BlinkML uses probabilistic guarantees for approximate model training.
result BlinkML can speed up ML training by 6.26x-629x with 95% prediction accuracy.
New sampling method improves search efficiency in machine learning.
problem Efficiently sampling effective solutions from large search spaces.
method Developed a parameterized family of coverage-based designs and algorithms for effective synthesis.
result Consistently outperforms existing exploratory sampling methods in sample mining and hyper-parameter optimization.
Experiment evaluates hospital case cost prediction models using Azure ML.
problem Accurate hospital case cost modelling for efficient financial management.
method Azure Machine Learning Studio tool for comparing 14 regression models.
result Robust regression, boosted decision tree, and decision forest models outperformed others.
Study optimizes CANN for actuarial tasks using RSM.
problem Optimizing hyperparameters for neural networks in actuarial science.
method Factorial design and response surface methodology (RSM).
result Reduced hyperparameter optimization from 288 to 188, achieving near-optimal performance.
MLSys aims to bridge ML and systems research.
problem Designing ML systems for real-world deployment is challenging.
method Foster a new conference and research community.
result MLSys conference focuses on intersection of systems and ML.
The rise of Big Data has led to new demands for Machine Learning (ML) systems to learn complex models with millions to billions of parameters, that promise adequate capacity to digest massive datasets and offer powerful predictive analytics thereupon. In order to run ML algorithms at such scales, on a distributed clust…
Machine learning combines data, model, and loss components.
problem Transforming human lives through ML applications.
method Combines data, model, and loss components in efficient implementations.
result Understanding ML as three components helps navigate its growing applications.
System detects overfitting in ML apps, improving quality and efficiency.
problem Overfitting in ML applications during continuous development.
method ease. ml/meter system for automated overfitting detection and measurement.
result Probabilistic overfitting signals for developers to take actions.
MLPerf benchmarks ML inference systems across diverse hardware.
problem Challenges in evaluating ML inference systems due to variety and complexity.
method MLPerf prescribes a set of rules and best practices for comparability.
result First call for submissions yielded over 600 reproducible measurements.
Geometrically proves twisted Poincaré duality for orientable Poisson manifolds.
problem Establishing twisted Poincaré duality for Poisson manifolds.
method Geometrically reinterprets algebraic constructions of twisted Poisson modules and Poisson chain complexes.
result Explicit chain isomorphism between Poisson cochain and chain complexes with coefficients in Poisson modules.
The abstract semiclassicalises quantum group principal bundles to Poisson geometry.
problem Semiclassicalising quantum group principal bundles to Poisson geometry.
method The theory is developed for Poisson manifolds with Poisson-compatible contravariant connections, and for Poisson-Lie groups with bicovariant Poisson-compatible contravariant connections.
result The construction of the Poisson level of the q-Hopf fibration and the spin connection on a principal bundle. The DoD needs a robust process to evaluate AI/ML model performance and robustness.
problem AI/ML models are brittle and nonrobust, posing risks in national security.
method Reviews AI/ML development process and best practices for evaluation.
result Recommendations for DoD evaluators to ensure robust AI/ML capabilities.
A review of ML and DL for ecological data analysis.
problem Understanding the strengths and limitations of ML and DL in ecological research.
method Historical overview, algorithm families, differences, universal principles, and emerging trends.
result ML and DL excel in prediction tasks but are still debated for causal inference.
We propose a Poisson-Lie analog of the symplectic induction procedure, using an appropriate Poisson generalization of the reduction of symplectic manifolds with symmetry. Having as basic tools the equivariant momentum maps of Poisson actions, the double group of a Poisson-Lie group and the reduction of Poisson manifold…
New Poisson structures on algebras linked to derivatives.
problem Understanding Poisson structures on trivial extension algebras.
method Introducing a correspondence between Poisson structures and derivatives, and formulating results in terms of Lie algebroids.
result One-to-one correspondence between Poisson structures and data involving derivatives.
New Lie groups found for Poisson diffeomorphisms.
problem Finding Lie group structures on Poisson diffeomorphism groups.
method Using Poisson groupoids, develop Lie group structures.
result Poisson diffeomorphism groups of various Poisson manifolds are regular Lie groups.
MLPerf benchmarks ML training to drive performance improvements.
problem Unique challenges in ML training benchmarks.
method Developed MLPerf to overcome ML training's specific challenges.
result Quantitatively evaluated MLPerf's effectiveness.
AI methods are energy-intensive, but efficiency alone isn't enough for sustainability.
problem AI methods are energy-intensive and contribute to climate change.
method Critically examines the limitations of efficiency in improving environmental sustainability of AI.
result Efficiency alone is insufficient to address the environmental impacts of AI.
A rigorous ML pipeline for binary classification in biomedical studies, focusing on pancreatic cancer.
problem Handling bias in ML models for complex biomedical data.
method Customizable ML analysis pipeline with 9 algorithms, hyperparameter optimization, and thorough evaluation.
result Comparison of ML algorithms to ExSTraCS, highlighting interpretability and bias handling.
We construct a Poisson isomorphism between the formal Poisson manifolds g^* and G^*, where g is a finite dimensional quasitriangular Lie bialgebra. Here g^* is equipped with its Lie-Poisson (or Kostant-Kirillov-Souriau) structure, and G^* with its Poisson-Lie structure. We also quantize Poisson-Lie dynamical r-matrices…
Research evaluates model extraction attacks on complex ML models and introduces a defense.
problem Model extraction attacks steal functionality of ML models through prediction APIs.
method Evaluation of Knockoff nets and introduction of a defense.
result Realistic adversaries can effectively steal complex ML models and evade known defenses.
This paper tackles hidden technical debts in fair ML systems for Fintech.
problem Building fair machine learning systems in financial services.
method Examining key stages of ML system development and deployment.
result Technical debts exist in deploying fair ML systems in Fintech.
The paper establishes a Poisson Poincaré-Dulac theorem for Poisson-flat connections.
problem Analyzing Poisson-flat connections with logarithmic poles.
method Defining an Euler-Poisson principal part and residue theory, establishing a Poisson Poincaré-Dulac theorem.
result Any logarithmic Poisson-flat connection is holomorphically gauge equivalent to a pure Euler-Poisson normal form.
The first cohomology of Poisson algebras is described and conditions for its vanishing are established.
problem Understanding the first cohomology of Poisson algebras.
method Description and mapping of first cohomology to intrinsic cohomologies of Poisson submanifolds, formulation of vanishing conditions.
result Necessary and sufficient conditions for the vanishing of the first cohomology of infinitesimal Poisson algebras are derived.
The paper normalizes Poisson saturation of coregular submanifolds.
problem Normalizing the Poisson saturation of coregular submanifolds.
method Normal form construction and Poisson geometry analysis.
result Local Poisson saturation of coregular submanifolds is an embedded Poisson submanifold with a normal form.
This paper emphasizes the need for uncertainty quantification in data-driven ML models for nuclear engineering.
problem Uncertainty in ML predictions due to data noise, model architecture, and stochastic training.
method Explains and compares uncertainties in physics-based and data-driven models, and presents techniques to quantify ML prediction uncertainties.
result The importance of uncertainty quantification in ML models for nuclear engineering applications.
Develops reduction method for strong Dirac maps.
problem Generalizing Poisson momentum maps.
method General procedure for reduction along strong Dirac maps.
result Recover and introduce new Poisson, quasi-Poisson, and Dirac reduced structures.
We reformulate the Poisson structure discovered by Fock and Rosly on moduli spaces of flat connections over marked surfaces in the framework of Poisson structures defined by Lie algebra actions and quasitriangular r-matrices, and we show that it is an example of a mixed product Poisson structure associated to pairs o…
A new Poisson bracket defined on Poisson structures with applications to fixed points and cohomology.
problem Defining a Poisson bracket on the space of Poisson structures.
method Constructing a Poisson bracket on P(M) depending on a volume form, and defining invariant of Poisson structures. result Invariant of Poisson structures detects unimodularity and related Poisson bracket for symplectic structures.
The study introduces a new equivalence for Poisson modules on complex projective varieties.
problem Understanding the structure of Poisson modules on complex projective varieties with mild singularities.
method Introducing a weak concept of Morita equivalence in the birational context for Poisson modules.
result Poisson modules are classified into three types based on their properties.
ABOUT ML aims to improve transparency in ML lifecycle documentation.
problem Lack of standard documentation in machine learning lifecycle.
method Initiative to operationalize ML transparency and standardize documentation.
result Helps address gaps in ML lifecycle documentation.
Abstract: Homotopy Poisson algebra models for reduced spaces derived from Poisson structures.
problem Homotopy Poisson algebra models for reduced spaces.
method Cattaneo-Zambon compatibility and regularity conditions, equivariant map, homotopy Poisson algebra.
result Derivation of homotopy Poisson algebra generalizing classical BFV algebra.
Study Poisson algebras for Hamiltonian systems linearization.
problem Linearize dynamics along Poisson submanifolds.
method Use contravariant derivative to characterize Poisson algebras.
result Infinitesimal Poisson algebras provide a framework for Hamiltonization.
Paper proposes using creative ML for game design.
problem Lack of creative ML in game design.
method Leverage existing creative ML systems for game content creation.
result Illustrates how creative ML can inform new game design systems.
The paper studies Poisson vector fields and tensor deformations.
problem Understanding Poisson vector fields and tensor deformations.
method Proving Poisson properties of lifts and describing infinitesimal deformations.
result Infinitesimal deformations of Poisson tensors have been described.
Gaussian surrogates improve Poisson imaging performance at low doses.
problem Improving Poisson imaging performance at low doses.
method Analysis of Poisson and Gaussian surrogate reconstruction objectives under Poisson noise.
result Gaussian surrogates can achieve MSE comparable to Poisson MAP at low doses.
This chapter is an attempt to present a mathematical theory of compound fractional Poisson processes. The chapter begins with the characterization of a well-known Lévy process: The compound Poisson process. The semi-Markov extension of the compound Poisson process naturally leads to the compound fractional Poisson proc…
Cohomology of 'book' Lie algebra Poisson structure computed.
problem Computing the cohomology of a specific Lie algebra structure.
method Direct computation of cohomology for the given Lie algebra structure.
result Explicit formula for Poisson cohomology of the 'book' Lie algebra.
Guidelines for using explainable ML to avoid misuse.
problem Misuse of explainable ML, especially for harmful purposes.
method Proposed guidelines to promote best practices.
result Promote interpretable models and testing methods.
Causal ML predicts treatment outcomes, aiding personalized medicine.
problem Predicting individualized treatment effects for personalized medicine.
method Flexible, data-driven methods using causal inference with clinical trial and real-world data.
result Causal ML allows for estimating individualized treatment effects.
ML-FFs use ML to bridge chem. accuracy and efficiency.
problem Narrowing the gap between ab initio and classical FFs.
method Learn potential energy from structure data without fixed bonds.
result ML-FFs can achieve accuracy of ab initio methods with classical efficiency.
We address the question of duality for the dynamical Poisson groupoids of Etingof and Varchenko over a contractible base. We also give an explicit description for the coboundary case associated with the solutions of the classical dynamical Yang-Baxter equation on simple Lie algebras as classified by the same authors. O…