PyTorchKR Papers

최신 논문

총 802편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2503.02655

Quantum Geometry insights in Deep Learning

In this paper, we explore the fundamental role of the Monge-Amp\`ere equation in deep learning, particularly in the context of Boltzmann machines and energy-based models. We first review the structure of Boltzmann learning and its relation to free energy minimization. We then establish a connection between optimal transport theory and deep learning, demonstrating how the Monge-Amp\`ere equation governs probability transformations in generative models. Additionally, we provide insights from quantum geometry, showing that the space of covariance matrices arising in the learning process coincides with the Connes-Araki-Haagerup (CAH) cone in von Neumann algebra theory. Furthermore, we introduce an alternative approach based on renormalization group (RG) flow, which, while distinct from the optimal transport perspective, reveals another manifestation of the Monge-Amp\`ere domain in learning dynamics. This dual perspective offers a deeper mathematical understanding of hierarchical feature learning, bridging concepts from statistical mechanics, quantum geometry, and deep learning theory.

arXiv:2503.01635

The Emergence of Grammar through Reinforcement Learning

The evolution of grammatical systems of syntactic and semantic composition is modeled here with a novel application of reinforcement learning theory. To test the functionalist thesis that speakers' expressive purposes shape their language, we include within the model a probability distribution over different messages that could be expressed in a given context. The proposed learning and production algorithm then breaks down language learning into a sequence of simple steps, such that each step benefits from the message probabilities. The results are presented in the form of numerical simulations of language histories and analytic proofs. The potential for applying these mathematical models to the study of natural language is illustrated with two case studies from the history of English.

arXiv:2502.16816

Finite-Sample Analysis of Policy Evaluation for Robust Average Reward Reinforcement Learning

We present the first finite-sample analysis for policy evaluation in robust average-reward Markov Decision Processes (MDPs). Prior works in this setting have established only asymptotic convergence guarantees, leaving open the question of sample complexity. In this work, we address this gap by establishing that the robust Bellman operator is a contraction under the span semi-norm, and developing a stochastic approximation framework with controlled bias. Our approach builds upon Multi-Level Monte Carlo (MLMC) techniques to estimate the robust Bellman operator efficiently. To overcome the infinite expected sample complexity inherent in standard MLMC, we introduce a truncation mechanism based on a geometric distribution, ensuring a finite constant sample complexity while maintaining a small bias that decays exponentially with the truncation level. Our method achieves the order-optimal sample complexity of $\tilde{\mathcal{O}}(\epsilon^{-2})$ for robust policy evaluation and robust average reward estimation, marking a significant advancement in robust reinforcement learning theory.

arXiv:2502.16759

The Blessing of Reasoning: LLM-Based Contrastive Explanations in Black-Box Recommender Systems

Modern recommender systems use ML models to predict consumer preferences from consumption history. Although these "black-box" models achieve impressive predictive performance, they often suffer from a lack of transparency and explainability. Contrary to the presumed tradeoff between explainability and accuracy, we show that integrating large language models (LLMs) with deep neural networks (DNNs) can improve both. We propose LR-Recsys, which augments DNN-based systems with LLM reasoning capabilities. LR-Recsys introduces a contrastive-explanation generator that produces human-readable positive explanations and negative explanations. These explanations are embedded via a fine-tuned autoencoder and combined with consumer and product features to improve predictions. Beyond offering explainability, we show that LR-Recsys also improves learning efficiency and predictive accuracy, as supported by high-dimensional, multi-environment statistical learning theory. LR-Recsys outperforms state-of-the-art recommender systems by 3-14% on three real-world datasets. Importantly, our analysis reveals that these gains primarily derive from LLMs' reasoning capabilities rather than their external domain knowledge. LR-RecSys presents an effective approach to combine LLMs with traditional DNNs, two of the most widely used ML models today. The explanations generated by LR-Recsys provide actionable insights for consumers, sellers, and platforms, helping to build trust, optimize product offerings, and inform targeting strategies.

arXiv:2502.14047

Towards a Learning Theory of Representation Alignment

It has recently been argued that AI models' representations are becoming aligned as their scale and performance increase. Empirical analyses have been designed to support this idea and conjecture the possible alignment of different representations toward a shared statistical model of reality. In this paper, we propose a learning-theoretic perspective to representation alignment. First, we review and connect different notions of alignment based on metric, probabilistic, and spectral ideas. Then, we focus on stitching, a particular approach to understanding the interplay between different representations in the context of a task. Our main contribution here is relating properties of stitching to the kernel alignment of the underlying representation. Our results can be seen as a first step toward casting representation alignment as a learning-theoretic problem.

arXiv:2502.13030

Likelihood-Ratio Regularized Quantile Regression: Adapting Conformal Prediction to High-Dimensional Covariate Shifts

We consider the problem of conformal prediction under covariate shift. Given labeled data from a source domain and unlabeled data from a covariate shifted target domain, we seek to construct prediction sets with valid marginal coverage in the target domain. Most existing methods require estimating the unknown likelihood ratio function, which can be prohibitive for high-dimensional data such as images. To address this challenge, we introduce the likelihood ratio regularized quantile regression (LR-QR) algorithm, which combines the pinball loss with a novel choice of regularization in order to construct a threshold function without directly estimating the unknown likelihood ratio. We show that the LR-QR method has coverage at the desired level in the target domain, up to a small error term that we can control. Our proofs draw on a novel analysis of coverage via stability bounds from learning theory. Our experiments demonstrate that the LR-QR algorithm outperforms existing methods on high-dimensional prediction tasks, including a regression task for the Communities and Crime dataset, and an image classification task from the WILDS repository.

arXiv:2502.08457

Learning Theory for Kernel Bilevel Optimization

Bilevel optimization has emerged as a technique for addressing a wide range of machine learning problems that involve an outer objective implicitly determined by the minimizer of an inner problem. In this paper, we investigate the generalization properties for kernel bilevel optimization problems where the inner objective is optimized over a Reproducing Kernel Hilbert Space. This setting enables rich function approximation while providing a foundation for rigorous theoretical analysis. In this context, we establish novel generalization error bounds for the bilevel problem under finite-sample approximation. Our approach adopts a functional perspective, inspired by (Petrulionyte et al., 2024), and leverages tools from empirical process theory and maximal inequalities for degenerate $U$-processes to derive uniform error bounds. These generalization error estimates allow to characterize the statistical accuracy of gradient-based methods applied to the empirical discretization of the bilevel problem.

arXiv:2502.09106

Scaling Law for Stochastic Gradient Descent in Quadratically Parameterized Linear Regression

In machine learning, the scaling law describes how the model performance improves with the model and data size scaling up. From a learning theory perspective, this class of results establishes upper and lower generalization bounds for a specific learning algorithm. Here, the exact algorithm running using a specific model parameterization often offers a crucial implicit regularization effect, leading to good generalization. To characterize the scaling law, previous theoretical studies mainly focus on linear models, whereas, feature learning, a notable process that contributes to the remarkable empirical success of neural networks, is regretfully vacant. This paper studies the scaling law over a linear regression with the model being quadratically parameterized. We consider infinitely dimensional data and slope ground truth, both signals exhibiting certain power-law decay rates. We study convergence rates for Stochastic Gradient Descent and demonstrate the learning rates for variables will automatically adapt to the ground truth. As a result, in the canonical linear regression, we provide explicit separations for generalization curves between SGD with and without feature learning, and the information-theoretical lower bound that is agnostic to parametrization method and the algorithm. Our analysis for decaying ground truth provides a new characterization for the learning dynamic of the model.

arXiv:2502.08597

Markets with Heterogeneous Agents: Dynamics and Survival of Bayesian vs. No-Regret Learners

We analyze the performance of heterogeneous learning agents in asset markets with stochastic payoffs. Our main focus is on comparing Bayesian learners and no-regret learners who compete in markets and identifying the conditions under which each approach is more effective. Surprisingly, we find that low regret is not sufficient for survival: an agent can have regret as low as $O(\log T)$ but still vanish when competing against a Bayesian with a finite prior and any positive prior probability on the correct model. On the other hand, we show that Bayesian learning is fragile, while no-regret learning requires less knowledge of the environment and is therefore more robust. Motivated by the strengths and weaknesses of both approaches, we propose a balanced strategy for utilizing Bayesian updates that improves robustness and adaptability to distribution shifts, providing a step toward a best-of-both-worlds learning approach. The method is general, efficient, and easy to implement. Finally, we formally establish the relationship between the notions of survival and market dominance studied in economics and the framework of regret minimization, thus bridging these theories. More broadly, our work contributes to the understanding of dynamics with heterogeneous types of learning agents and their impact on markets.

arXiv:2502.05300

Parameter Symmetry Potentially Unifies Deep Learning Theory

The dynamics of learning in modern large AI systems is hierarchical, often characterized by abrupt, qualitative shifts akin to phase transitions observed in physical systems. While these phenomena hold promise for uncovering the mechanisms behind neural networks and language models, existing theories remain fragmented, addressing specific cases. In this position paper, we advocate for the crucial role of the research direction of parameter symmetries in unifying these fragmented theories. This position is founded on a centralizing hypothesis for this direction: parameter symmetry breaking and restoration are the unifying mechanisms underlying the hierarchical learning behavior of AI models. We synthesize prior observations and theories to argue that this direction of research could lead to a unified understanding of three distinct hierarchies in neural networks: learning dynamics, model complexity, and representation formation. By connecting these hierarchies, our position paper elevates symmetry -- a cornerstone of theoretical physics -- to become a potential fundamental principle in modern AI.

arXiv:2502.03504

Immersion for AI: Immersive Learning with Artificial Intelligence

This work reflects upon what Immersion can mean from the perspective of an Artificial Intelligence (AI). Applying the lens of immersive learning theory, it seeks to understand whether this new perspective supports ways for AI participation in cognitive ecologies. By treating AI as a participant rather than a tool, it explores what other participants (humans and other AIs) need to consider in environments where AI can meaningfully engage and contribute to the cognitive ecology, and what the implications are for designing such learning environments. Drawing from the three conceptual dimensions of immersion - System, Narrative, and Agency - this work reinterprets AIs in immersive learning contexts. It outlines practical implications for designing learning environments where AIs are surrounded by external digital services, can interpret a narrative of origins, changes, and structural developments in data, and dynamically respond, making operational and tactical decisions that shape human-AI collaboration. Finally, this work suggests how these insights might influence the future of AI training, proposing that immersive learning theory can inform the development of AIs capable of evolving beyond static models. This paper paves the way for understanding AI as an immersive learner and participant in evolving human-AI cognitive ecosystems.

arXiv:2502.05098

Learning Temporal Invariance in Android Malware Detectors

Learning-based Android malware detectors degrade over time due to natural distribution drift caused by malware variants and new families. This paper systematically investigates the challenges classifiers trained with empirical risk minimization (ERM) face against such distribution shifts and attributes their shortcomings to their inability to learn stable discriminative features. Invariant learning theory offers a promising solution by encouraging models to generate stable representations crossing environments that expose the instability of the training set. However, the lack of prior environment labels, the diversity of drift factors, and low-quality representations caused by diverse families make this task challenging. To address these issues, we propose TIF, the first temporal invariant training framework for malware detection, which aims to enhance the ability of detectors to learn stable representations across time. TIF organizes environments based on application observation dates to reveal temporal drift, integrating specialized multi-proxy contrastive learning and invariant gradient alignment to generate and align environments with high-quality, stable representations. TIF can be seamlessly integrated into any learning-based detector. Experiments on a decade-long dataset show that TIF excels, particularly in early deployment stages, addressing real-world needs and outperforming state-of-the-art methods.

arXiv:2502.00976

Learning the Integral Quadratic Constraints on Plant-Model Mismatch

While a characterization of plant-model mismatch is necessary for robust control, the mismatch usually can not be described accurately due to the lack of knowledge about the plant model or the complexity of nonlinear plants. Hence, this paper considers this problem in a data-driven way, where the mismatch is captured by parametric forms of integral quadratic constraints (IQCs) and the parameters contained in the IQC equalities are learned from sampled trajectories from the plant. To this end, a one-class support vector machine (OC-SVM) formulation is proposed, and its generalization performance is analyzed based on the statistical learning theory. The proposed approach is demonstrated by a single-input-single-output time delay mismatch and a nonlinear two-phase reactor with a linear nominal model, showing accurate recovery of frequency-domain uncertainties.

arXiv:2502.00607

PAC Learning is just Bipartite Matching (Sort of)

The main goal of this article is to convince you, the reader, that supervised learning in the Probably Approximately Correct (PAC) model is closely related to -- of all things -- bipartite matching! En-route from PAC learning to bipartite matching, I will overview a particular transductive model of learning, and associated one-inclusion graphs, which can be viewed as a generalization of some of the hat puzzles that are popular in recreational mathematics. Whereas this transductive model is far from new, it has recently seen a resurgence of interest as a tool for tackling deep questions in learning theory. A secondary purpose of this article could be as a (biased) tutorial on the connections between the PAC and transductive models of learning.

arXiv:2501.17115

Evidence on the Regularisation Properties of Maximum-Entropy Reinforcement Learning

The generalisation and robustness properties of policies learnt through Maximum-Entropy Reinforcement Learning are investigated on chaotic dynamical systems with Gaussian noise on the observable. First, the robustness under noise contamination of the agent's observation of entropy regularised policies is observed. Second, notions of statistical learning theory, such as complexity measures on the learnt model, are borrowed to explain and predict the phenomenon. Results show the existence of a relationship between entropy-regularised policy optimisation and robustness to noise, which can be described by the chosen complexity measures.

arXiv:2501.14474

The Pseudo-Dimension of Contracts

Algorithmic contract design studies scenarios where a principal incentivizes an agent to exert effort on her behalf. In this work, we focus on settings where the agent's type is drawn from an unknown distribution, and formalize an offline learning framework for learning near-optimal contracts from sample agent types. A central tool in our analysis is the notion of pseudo-dimension from statistical learning theory. Beyond its role in establishing upper bounds on the sample complexity, pseudo-dimension measures the intrinsic complexity of a class of contracts, offering a new perspective on the tradeoffs between simplicity and optimality in contract design. Our main results provide essentially optimal tradeoffs between pseudo-dimension and representation error (defined as the loss in principal's utility) with respect to linear and bounded contracts. Using these tradeoffs, we derive sample- and time-efficient learning algorithms, and demonstrate their near-optimality by providing almost matching lower bounds on the sample complexity. Conversely, for unbounded contracts, we prove an impossibility result showing that no learning algorithm exists. Finally, we extend our techniques in three important ways. First, we provide refined pseudo-dimension and sample complexity guarantees for the combinatorial actions model, revealing a novel connection between the number of critical values and sample complexity. Second, we extend our results to menus of contracts, showing that their pseudo-dimension scales linearly with the menu size. Third, we adapt our algorithms to the online learning setting, where we show that, a polynomial number of type samples suffice to learn near-optimal bounded contracts. Combined with prior work, this establishes a formal separation between expert advice and bandit feedback for this setting.

arXiv:2501.12747

Singular leaning coefficients and efficiency in learning theory

Singular learning models with non-positive Fisher information matrices include neural networks, reduced-rank regression, Boltzmann machines, normal mixture models, and others. These models have been widely used in the development of learning machines. However, theoretical analysis is still in its early stages. In this paper, we examine learning coefficients, which indicate the general learning efficiency of deep linear learning models and three-layer neural network models with ReLU units. Finally, we extend the results to include the case of the Softmax function.

arXiv:2501.12737

Stability and Generalization of Quantum Neural Networks

Quantum neural networks (QNNs) play an important role as an emerging technology in the rapidly growing field of quantum machine learning. While their empirical success is evident, the theoretical explorations of QNNs, particularly their generalization properties, are less developed and primarily focus on the uniform convergence approach. In this paper, we exploit an advanced tool in statistical learning theory, i.e., algorithmic stability, to study the generalization of QNNs. We first establish high-probability generalization bounds for QNNs via uniform stability. Our bounds shed light on the key factors influencing the generalization performance of QNNs and provide practical insights into both the design and training processes. We next explore the generalization of QNNs on near-term noisy intermediate-scale quantum (NISQ) devices, highlighting the potential benefits of quantum noise. Moreover, we argue that previous analysis characterizes worst-case generalization guarantees, and we establish a refined optimization-dependent generalization bound for QNNs via on-average stability. Numerical experiments on various real-world datasets support our theoretical findings.

arXiv:2501.10077공식 코드 · Papers with Code

Double descent in quantum machine learning

The double descent phenomenon challenges traditional statistical learning theory by revealing scenarios where larger models do not necessarily lead to reduced performance on unseen data. While this counterintuitive behavior has been observed in a variety of classical machine learning models, particularly modern neural network architectures, it remains elusive within the context of quantum machine learning. In this work, we analytically demonstrate that quantum learning models can exhibit double descent behavior by drawing on insights from linear regression and random matrix theory. Additionally, our numerical experiments on quantum kernel methods across different real-world datasets and system sizes further confirm the existence of a test error peak, a characteristic feature of double descent. Our findings provide evidence that quantum models can operate in the modern, overparameterized regime without experiencing overfitting, thereby opening pathways to improved learning performance beyond traditional statistical learning theory.

arXiv:2501.09189

Testing Noise Assumptions of Learning Algorithms

We pose a fundamental question in computational learning theory: can we efficiently test whether a training set satisfies the assumptions of a given noise model? This question has remained unaddressed despite decades of research on learning in the presence of noise. In this work, we show that this task is tractable and present the first efficient algorithm to test various noise assumptions on the training data. To model this question, we extend the recently proposed testable learning framework of Rubinfeld and Vasilyan (2023) and require a learner to run an associated test that satisfies the following two conditions: (1) whenever the test accepts, the learner outputs a classifier along with a certificate of optimality, and (2) the test must pass for any dataset drawn according to a specified modeling assumption on both the marginal distribution and the noise model. We then consider the problem of learning halfspaces over Gaussian marginals with Massart noise (where each label can be flipped with probability less than $1/2$ depending on the input features), and give a fully-polynomial time testable learning algorithm. We also show a separation between the classical setting of learning in the presence of structured noise and testable learning. In fact, for the simple case of random classification noise (where each label is flipped with fixed probability $\eta = 1/2$), we show that testable learning requires super-polynomial time while classical learning is trivial.