This work analyzes tree-based methods from a ranking perspective, providing insights and new statistics.
problem Understanding the effectiveness of tree-based methods in finite-sample settings, especially symbolic feature selection.
method Local ranking perspective, finite-sample analysis, oracle bounds, posterior contraction results, concordant divergence statistics.
result New insights and statistics for evaluating symbolic feature mappings.
Learning representations of data is an important problem in statistics and machine learning. While the origin of learning representations can be traced back to factor analysis and multidimensional scaling in statistics, it has become a central theme in deep learning with important applications in computer vision and co…
Teaches deep learning to statisticians.
problem Statisticians lack expertise in deep learning.
method Developed a program and taught DL to statistics graduate students.
result Provided tips and resources for teaching DL.
Socio-economic inequalities are manifested in different aspects of our social life. We discuss various aspects, beginning with the evolutionary and historical origins, and discussing the major issues from the social and economic point of view. The subject has attracted scholars from across various disciplines, includin…
We consider statistical and algorithmic aspects of solving large-scale least-squares (LS) problems using randomized sketching algorithms. Prior results show that, from an \emph{algorithmic perspective}, when using sketching matrices constructed from random projections and leverage-score sampling, if the number of sampl…
Unified tutorial on AMP for high-dimensional problems.
problem Structured high-dimensional statistical problems.
method Statistical perspective of AMP and its applications.
result Unified and strengthened results in AMP literature.
Researchers explore statistical perspectives to understand GNN generalization.
problem Limited mathematical understanding of GNN performance.
method Three broad frameworks: learning theory, asymptotics, and random graph models.
result Various theoretical results and open questions identified.
Deep learning's success is puzzling from a statistical perspective.
problem Deep learning's success is puzzling from a statistical perspective.
method Physics-informed investigation of deep learning features and surprises.
result Neural scaling laws and their interplay with inductive biases.
Statistical analysis reveals pitfalls in climate network construction.
problem Uncertainty in correlation estimates affects network characteristics.
method Simulated isotropic random fields and resampling procedures.
result Spurious link bundles and high-degree clusters are common.
We consider statistical as well as algorithmic aspects of solving large-scale least-squares (LS) problems using randomized sketching algorithms. For a LS problem with input data (X,Y)∈Rn×p×Rn, sketching algorithms use a sketching matrix, S∈Rr×n with $r \…
Database theory and database practice are typically the domain of computer scientists who adopt what may be termed an algorithmic perspective on their data. This perspective is very different than the more statistical perspective adopted by statisticians, scientific computers, machine learners, and other who work on wh…
Survey on statistical learning theory for control, focusing on linear systems.
problem Applying machine learning techniques to control systems, especially linear ones.
method Adapting tools from modern high-dimensional statistics and learning theory.
result Recent advances in statistical learning theory for control, particularly for linear systems.
The statistical analysis of discrete data has been the subject of extensive statistical research dating back to the work of Pearson. In this survey we review some recently developed methods for testing hypotheses about high-dimensional multinomials. Traditional tests like the χ2 test and the likelihood ratio test ca…
Unified approach for quantum and classical learning from evaluation oracles.
problem Learning from evaluation oracles in quantum and classical settings.
method Inspired by Kearns' SQ and Valiant's weak evaluation oracle, a unified framework is established.
result Characterizes query complexity for learning linear function classes and extends learnability results for quantum circuits.
Survey on factor models and their applications in econometrics.
problem Estimating low-rank structures in high-dimensional models.
method Low-rank recovery techniques for factor model estimation.
result New insights into factor model applications in econometrics.
In recent years, ideas from statistics and scientific computing have begun to interact in increasingly sophisticated and fruitful ways with ideas from computer science and the theory of algorithms to aid in the development of improved worst-case algorithms that are useful for large-scale scientific and Internet data an…
Survey of robust streaming techniques and their relationships.
problem Challenges in robust streaming and online learning.
method Overview and survey of robust streaming techniques, unifying theorems.
result Proved the relationship between robust streaming techniques.
Review of Gerber-Shiu function for practical actuarial science.
problem Difficulty in numerical approximation and statistical inference of Gerber-Shiu function.
method Comprehensive review of formulations, surplus processes, numerical methods, and statistical inference.
result Enhanced understanding and practical guide for Gerber-Shiu function.
New approach to convex hulls for low-rank problems.
problem Characterizing convex hulls for low-rank sets.
method Matrix perspective function and orthogonal projection matrices.
result Strong relaxations for various low-rank problems.
How should statistical procedures be designed so as to be scalable computationally to the massive datasets that are increasingly the norm? When coupled with the requirement that an answer to an inferential question be delivered within a certain time budget, this question has significant repercussions for the field of s…
We discuss social network analysis from the perspective of economics. We organize the presentaion around the theme of externalities: the effects that one's behavior has on others' well-being. Externalities underlie the interdependencies that make networks interesting. We discuss network formation, as well as interactio…
This work analyzes CoT prompting methods from a statistical estimation perspective.
problem Improving the effectiveness of LLMs in solving multi-step reasoning problems.
method Introducing a multi-step latent variable model to characterize CoT prompting from a statistical estimation viewpoint.
result The CoT estimator is equivalent to a Bayesian estimator when the pretraining dataset is large.
One popular method for dealing with large-scale data sets is sampling. For example, by using the empirical statistical leverage scores as an importance sampling distribution, the method of algorithmic leveraging samples and rescales rows/columns of data matrices to reduce the data size before performing computations on…
We consider solution of stochastic storage problems through regression Monte Carlo (RMC) methods. Taking a statistical learning perspective, we develop the dynamic emulation algorithm (DEA) that unifies the different existing approaches in a single modular template. We then investigate the two central aspects of regres…
Distillation improves simple models by approximating complex labels.
problem Why does distillation improve simple models?
method Statistical perspective on distillation, connecting to extreme multiclass retrieval.
result Distillation helps by approximating underlying class-probabilities, reducing bias and variance.
RLHF uses human feedback to train AI models, posing statistical challenges.
problem Aligning AI models with human preferences using noisy, subjective feedback.
method Supervised fine-tuning, reward modeling, policy optimization, statistical ideas.
result Statistical methods for reward function learning and policy optimization.
This paper advances FL algorithms for composite optimization and statistical recovery.
problem Federated learning optimization and statistical recovery in composite settings.
method Proposes Fast Federated Dual Averaging for strongly convex and smooth loss, and Multi-stage Federated Dual Averaging for restricted strongly convex and smooth loss.
result Establishes state-of-the-art iteration and communication complexity, and high probability complexity bound with linear speedup.
Study trade-offs between statistical and computational efficiency in variational inference.
problem Optimizing statistical accuracy vs. computational efficiency in Bayesian inference.
method Case study on Gaussian inferential models with diagonal plus low-rank precision matrices, analyzing Bayesian posterior inference and frequentist uncertainty quantification errors.
result Lower-rank models reduce variance and accelerate convergence but increase posterior inference error.
New research shows LLMs can't be explained by statistical generalization alone.
problem Understanding why large language models (LLMs) perform well despite statistical generalization limitations.
method Examined the non-identifiability of AR probabilistic models and their implications for LLMs.
result Non-identifiability of LLMs leads to different behaviors and requires a separate theoretical explanation.
Improved privacy-preserving statistical estimates with customizable noise reduction.
problem Balancing privacy and accuracy in statistical estimation.
method Introducing the Brownian mechanism, which adds Gaussian noise to a sequence of estimates, gradually reducing it based on the practitioner's needs.
result The Brownian mechanism produces more accurate estimates while maintaining strong privacy guarantees, outperforming existing methods.
Deep learning has sparked a network of mutual interactions between different disciplines and AI. Naturally, each discipline focuses and interprets the workings of deep learning in different ways. This diversity of perspectives on deep learning, from neuroscience to statistical physics, is a rich source of inspiration t…
Breiman discusses two statistical cultures, advocating for more research on 'before' and 'after' the black box.
problem Statistical modeling lacks exploration of processes before and after the 'black box'.
method Analyzes Breiman's visual metaphor of two statistical cultures.
result Promotes the importance of studying the 'before' and 'after' of data transformations.
The paper establishes concentration bounds for embeddings of generative models.
problem Analyzing statistical properties of generative models.
method High probability concentration bounds on sample vector embeddings using Data Kernel Perspective Space.
result Determines the number of samples needed for accurate approximation of generative model embeddings.
Paper questions RNN and LSTM's long-term memory and introduces a new definition.
problem Whether RNN and LSTM have long-term memory.
method Introduced a new definition of long-term memory and modified RNN and LSTM to test it.
result RNN and LSTM do not meet the new definition of long-term memory.
A theory of deep learning is emerging, focusing on training dynamics and statistics.
problem Develop a scientific theory to understand deep learning.
method Synthesize research into five areas: idealized settings, tractable limits, mathematical laws, hyperparameters, and universal behaviors.
result The emerging theory is a mechanics of the learning process, named learning mechanics.
Lecture notes on reinforcement learning using statistical methods.
problem Addressing the exploration-exploitation dilemma in decision making.
method Frequentist and Bayesian approaches, function approximation, neural networks.
result Unified framework for decision making and estimation.
Develops a new theory of loss functions for statistical machine learning.
problem Evaluation of solutions in binary and multiclass classification problems.
method Defines loss functions as subgradients of support functions of convex sets, enabling a calculus of losses.
result Provides a novel perspective on losses and develops a calculus that interpolates between different losses.
Since their emergence in the 1990's, the support vector machine and the AdaBoost algorithm have spawned a wave of research in statistical machine learning. Much of this new research falls into one of two broad categories: kernel methods and ensemble methods. In this expository article, I discuss the main ideas behind t…
The study analyzes the conflict between group fairness and individual fairness in machine learning.
problem The conflict between group fairness (optimal statistical parity) and individual fairness in machine learning.
method Established sufficient conditions for the compatibility between optimal statistical parity and individual fairness requirements.
result Identified regions along the Pareto frontier that satisfy individual fairness requirements.
We present a novel method for frequentist statistical inference in M-estimation problems, based on stochastic gradient descent (SGD) with a fixed step size: we demonstrate that the average of such SGD sequences can be used for statistical inference, after proper scaling. An intuitive analysis using the Ornstein-Uhlen…
Injectivity of ReLU networks studied using statistical physics.
problem When can the input of a ReLU neural network be inferred from its output?
method Connection to spherical integral geometry and statistical physics.
result Replica symmetry-breaking theory and Gordon's min--max theorem provide insights into the injectivity threshold.
Machine fairness is impossible to achieve fully due to historical biases.
problem Machine learning models inherit biases from historical data, making it impossible to satisfy fairness metrics simultaneously.
method Presented a causal perspective to the impossibility theorem of fairness.
result It is impossible to satisfy fairness metrics like demographic parity, equal opportunity, and equalized odds simultaneously.
PILAF optimizes reward models from human feedback for better policy alignment.
problem Creating accurate reward models from human feedback for policy optimization.
method Policy-Interpolated Learning for Aligned Feedback (PILAF) that explicitly aligns preference learning with maximizing underlying oracle reward.
result PILAF is optimal from both optimization and statistical perspectives, demonstrating strong performance in RLHF settings.
New method detects uncertainty in neural networks for out-of-distribution detection.
problem Detecting out-of-distribution inputs to ensure model reliability.
method Predictive topological uncertainty (pTU) based on persistent homology.
result pTU provides a statistical framework for OOD detection.
A brief review is given of the minority game, an idealized model stimulated by a market of speculative agents, and its complex many-body behaviour. Particular consideration is given to analytic results for the model rather than discussions of its relevance in real-world situations.
PAQ8 is an open source lossless data compression algorithm that currently achieves the best compression rates on many benchmarks. This report presents a detailed description of PAQ8 from a statistical machine learning perspective. It shows that it is possible to understand some of the modules of PAQ8 and use this under…
Differential privacy is a statistical concept that can be explained through hypothesis testing.
problem Formalizing differential privacy as a statistical concept.
method Using David Blackwell's informativeness theorem, the paper shows differential privacy can be understood through hypothesis testing.
result The definition of f-differential privacy provides a unified framework for analyzing privacy bounds. Lectures on deep learning from a learning theory perspective.
problem Understanding how deep learning architectures lead to inductive bias.
method Statistical learning theory and stochastic optimization.
result Gradient descent on linear diagonal networks can lead to various forms of implicit bias.