Invariant Causal Set Covering Machines avoid spurious associations.
problem Learning algorithms for rule-based models are vulnerable to spurious associations.
method Building on invariant causal prediction, propose Invariant Causal Set Covering Machines for conjunctions/disjunctions of binary-valued rules.
result The method can identify causal parents of a variable of interest in polynomial time.
CDER uses Cover Trees and entropy to find labeled pointcloud differences.
problem Finding differences between labeled pointclouds.
method Combines Cover Trees and entropy to identify differences in labeled pointclouds.
result CDER identifies differences in labeled pointclouds in linear time.
We improve adversarial robustness calibration analysis for broader hypothesis sets.
problem Improving calibration for adversarial robustness in machine learning.
method A finer definition of calibration for adversarial robustness.
result Our results cover most common hypothesis sets in machine learning.
Combines TSP and SC to solve real-world vaccine distribution.
problem Combining TSP and SC for efficient vaccine distribution.
method Mixed Integer Programming (MIP) and machine learning.
result Machine learning approach improves solution efficiency.
The increased affordability of whole genome sequencing has motivated its use for phenotypic studies. We address the problem of learning interpretable models for discrete phenotypes from whole genomes. We propose a general approach that relies on the Set Covering Machine and a k-mer representation of the genomes. We sho…
This paper explains how model invariance improves generalization using data transformations.
problem Understanding why model invariance leads to better generalization performance.
method Introducing sample cover induced by transformations and refining generalization bounds.
result The sample covering number can be used to evaluate and select suitable data transformations.
Machine teaches IRL with minimal demonstrations.
problem Finding the minimum set of demonstrations for IRL.
method Formalized as a machine teaching problem, reduced to set cover, approximated efficiently.
result Efficient algorithm for determining maximally informative demonstrations.
Lecture notes on advanced linear regression methods.
problem Understanding the properties of linear regression estimators in high dimensions.
method Proposition-proof exploration of least squares, ridgeless, ridge, and lasso estimators.
result Detailed analysis of the existence, uniqueness, relations, computation, and non-asymptotic properties of these estimators.
The paper debiases machine learning predictions to correct bias in regression coefficients.
problem Bias in regression coefficients from machine learning predictions.
method Proposes an adversarial machine learning algorithm to de-bias predictions.
result Adversarial predictions recover true coefficients, while naive predictions are biased.
Lecture note for data science students on machine learning basics and advanced topics.
problem No specific problem stated; preparing students for advanced machine learning.
method Introduction of basic machine learning concepts and advanced topics.
result Students are prepared to study advanced machine learning topics.
Teaches uncertainty in ML through practical examples.
problem Lack of uncertainty teaching in ML curricula.
method Developed a curriculum and use cases.
result Motivates adoption of uncertainty concepts in AI courses.
Machine learning aids epidemiologists in analyzing big data.
problem Handling large, complex data in epidemiology.
method Explains principles and methods of supervised and unsupervised learning.
result Develops strategies for model evaluation and hyperparameter optimization.
Researchers use interpretable classifiers to predict antibiotic resistance.
problem Predicting antibiotic resistance from genome sequences.
method Set Covering Machines for highly interpretable models.
result Highly interpretable models for antibiotic resistance prediction.
Support vector classifier constructs confidence sets for binary classification.
problem Learning confidence sets with specific probability guarantees for binary classification.
method Support vector classifier to construct confidence sets.
result The proposed learner controls non-coverage rates and minimizes ambiguity with high probability.
Paper sets minimax bounds for Wasserstein distribution estimation.
problem Estimating a probability distribution using Wasserstein distance.
method Uses metric properties and weak moment assumptions.
result Upper and lower bounds on statistical minimax rates.
VR methods improve SGD for faster machine learning.
problem Efficiency in stochastic optimization for machine learning.
method Variance reduction techniques for stochastic optimization.
result VR methods achieve faster convergence than SGD.
We introduce a new nearest-prototype classifier, the prototype vector machine (PVM). It arises from a combinatorial optimization problem which we cast as a variant of the set cover problem. We propose two algorithms for approximating its solution. The PVM selects a relatively small number of representative points which…
Taxonomy for ML in simulations, covering patterns and algorithms.
problem Enhancing simulations using machine learning.
method Presentation of eight patterns and three algorithmic areas.
result Catalog of activities and patterns for ML integration in simulations.
New CFNN architecture approximates functions with machine accuracy.
problem Function approximation with high precision.
method Chebyshev Feature Neural Network (CFNN) with learnable frequencies.
result Achieves machine accuracy in function approximation.
Optimizes machine learning and system identification for real-world physical systems.
problem Estimating parameters in complex, real-world physical systems.
method Combines classical system identification and modern machine learning techniques using optimization-based approaches.
result Developed regularization strategies to incorporate prior knowledge into flexible models.
Machine learning selects the best prediction rules from noisy data.
problem Selection under uncertainty in machine learning.
method Statistical tools and inequalities to control noise in empirical estimates.
result Theoretical guarantees on selection outcomes under uncertainty.
This paper reviews various sampling methods from statistics and machine learning.
problem Addressing sampling methods in statistics and machine learning.
method Explains and reviews simple random sampling, bootstrapping, stratified sampling, cluster sampling, multistage sampling, network sampling, snowball sampling, and sampling from cumulative distribution function.
result Summarizes characteristics, pros, and cons of different sampling methods.
New method uses weak labels to create valid confidence sets for predictions.
problem Lack of labeled data in machine learning models.
method Developed a conformal prediction framework to provide valid predictive confidence sets using weakly labeled data.
result New coverage definition allows for tighter and more informative (but valid) confidence sets.
Surveying machine learning methods for economic forecasting.
problem Improving accuracy of economic forecasts using machine learning.
method Nowcasting, textual data, panel and tensor data, high-dimensional Granger causality tests, time series cross-validation, classification with economic losses.
result Recent advances in machine learning methods enhance economic forecasting accuracy.
Machine learning improves cloud cover forecasting.
problem Improving accuracy of total cloud cover predictions.
method Investigated multilayer perceptron, gradient boosting machines, random forest, logistic regression models.
result RF models provide the smallest increase in predictive performance, while MLP, POLR, and GBM approaches perform best.
Tackles bridging machine learning and control theory for safety-critical systems.
problem Ensuring reliability and safety in machine learning applications for safety-critical systems.
method Review of recent advances in learning and control theory, historical context.
result Importance of control theorists joining the conversation on learning-related problems.
Proper branched coverings are homeomorphisms on 3D balls or when branch set is empty.
problem Global injectivity of proper branched coverings on Euclidean balls.
method Analyzing the global injectivity of proper branched coverings defined on the Euclidean n-ball. result Proper branched coverings are homeomorphisms on 3D balls or when branch set is empty.
Bayesian framework improves ML classification models' uncertainty estimates.
problem Ensuring trustworthy AI predictions with explicit uncertainty quantification.
method Proposes a Bayesian framework for generative ML classification models that accounts for input measurement uncertainty.
result The BQDA model outperforms other models in terms of interpretability, explicit uncertainty modeling, and computational efficiency.
Paper introduces a novel method to generate diverse inputs for neural programming by example.
problem Synthesizing programs from input/output pairs using machine learning.
method Uses an SMT solver to generate diverse input-output pairs.
result Generated inputs improve model performance and generalization.
Curious examples of lifting spaces not as inverse limits of covering spaces.
problem Understanding inverse limits of covering spaces and their properties.
method Analyzing inverse limits of sequences of covering spaces over a given space.
result Presented examples of lifting spaces that cannot be obtained as inverse limits of covering spaces.
A subset of the sphere is said short if it is contained in an open hemisphere. A short closed set which is geodesically convex is called a cap. The following theorem holds: 1. The minimal number of short closed sets covering the n-sphere is n+2. 2. If n+2 short closed sets cover the n-sphere then (i) their inte…
New covering moves for 3-manifolds up to degree 4.
problem Relating colored link diagrams in 3-manifolds.
method Complete set of covering moves on braids in fixed degree d≥4. result Two local tangle replacements are sufficient after stabilization to the same degree at least 4.
Open and discrete maps with specific branch set images are equivalent to PL branched covers.
problem Understanding the equivalence of open and discrete maps and PL branched covers.
method Demonstrated that an open and discrete map f:SnoSn with a specific branch set image is equivalent to a PL branched cover up to homeomorphism. result Open and discrete maps with a specific branch set image are equivalent to PL branched covers.
Gromov-Thurston covers have Betti numbers as expected.
problem Understanding Betti numbers of branched covers of hyperbolic manifolds.
method Analyzing Gromov-Thurston branched covers and their Betti numbers.
result Betti numbers match expectations for non-divisible degree covers.
We present a new application and covering number bound for the framework of "Machine Learning with Operational Costs (MLOC)," which is an exploratory form of decision theory. The MLOC framework incorporates knowledge about how a predictive model will be used for a subsequent task, thus combining machine learning with t…
Proposes a new benchmark for deep neural network training.
problem Limited focus on DNN training efficiency.
method Develops TBD benchmark covering various applications and frameworks.
result Highlights inefficiencies in DNN training across different models and hardware.
A tutorial on neural machine translation and sequence-to-sequence models.
problem Handling human language through modeling sequential data.
method Explains and delves into neural networks and natural language processing techniques.
result Powerful tools for modeling sequential data in natural language.
Book introduces ML and AI for causal inference.
problem Uncertainty in causal relationships.
method Structural equation models, DAGs, SCMs, and Double/Debiased Machine Learning.
result Improved inference in causal models using predictive tools.
Study examines AI's role in robo-investing, focusing on benefits for specific investors.
problem Understanding the benefits of robo-investing for different investor types.
method Used a unique data set of brokerage accounts, analyzed various robo-investing strategies, compared human vs. machine performance.
result AI can provide benefits to low-income and high-risk-averse investors.
A new process model for machine learning applications with quality assurance.
problem Lack of standard process model for machine learning applications.
method Six-phase process model with quality assurance methodology.
result Proposes a new process model for machine learning applications.
In this paper we consider completed coverings that are branched coverings in the sense of Fox. For completed coverings between PL manifolds we give a characterization of the existence of a monodromy representation and the existence of a locally compact monodromy representation. These results stem from a characterizatio…
The Set Covering Machine (SCM) is a greedy learning algorithm that produces sparse classifiers. We extend the SCM for datasets that contain a huge number of features. The whole genetic material of living organisms is an example of such a case, where the number of feature exceeds 10^7. Three human pathogens were used to…
Pen-and-paper exercises cover various machine learning topics.
problem None explicitly stated, focuses on learning through exercises.
method Pen-and-paper exercises on machine learning topics.
result Comprehensive coverage of machine learning concepts through exercises.
Study covers of sphere with homeomorphisms lifting property.
problem Finite abelian covers of sphere with lifting homeomorphisms.
method Completely determined covers with specific lifting property.
result Properties of finite abelian covers with lifting homeomorphisms.
The paper studies liftable mapping class groups of cyclic covers of spheres.
problem Understanding liftable mapping class groups of cyclic covers of spheres.
method Derived finite generating sets, provided algorithms, determined isomorphism classes, derived presentations, and calculated normalizers and centralizers.
result Presentations and isomorphism classes of liftable mapping class groups for various covers.
Study stabilizes components of Galois cover moduli spaces.
problem Decide equivalence and stable equivalence of monodromy maps.
method Develop algebraic framework to study equivalence classes of monodromy maps.
result Recover a homological invariant that distinguishes equivalence classes.
Estimates open sets for fibrations, leading to volume vanishing results.
problem Estimating open sets for fibrations.
method Straightforward estimate for open sets with fundamental group constraints.
result Vanishing results for simplicial volume and minimal volume entropy for certain mapping tori.
In this paper we study the homeomorphisms of the disk that are liftable with respect to a simple branched covering. Since any such homeomorphism maps the branch set of the covering onto itself and liftability is invariant up to isotopy fixing the branch set, we are dealing in fact with liftable braids. We prove that th…