PyTorchKR Papers

최신 논문

총 516편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2407.06985공식 코드 · Papers with Code

PEER: Expertizing Domain-Specific Tasks with a Multi-Agent Framework and Tuning Methods

In domain-specific applications, GPT-4, augmented with precise prompts or Retrieval-Augmented Generation (RAG), shows notable potential but faces the critical tri-lemma of performance, cost, and data privacy. High performance requires sophisticated processing techniques, yet managing multiple agents within a complex workflow often proves costly and challenging. To address this, we introduce the PEER (Plan, Execute, Express, Review) multi-agent framework. This systematizes domain-specific tasks by integrating precise question decomposition, advanced information retrieval, comprehensive summarization, and rigorous self-assessment. Given the concerns of cost and data privacy, enterprises are shifting from proprietary models like GPT-4 to custom models, striking a balance between cost, security, and performance. We developed industrial practices leveraging online data and user feedback for efficient model tuning. This study provides best practice guidelines for applying multi-agent systems in domain-specific problem-solving and implementing effective agent tuning strategies. Our empirical studies, particularly in the financial question-answering domain, demonstrate that our approach achieves 95.0% of GPT-4's performance, while effectively managing costs and ensuring data privacy.

arXiv:2407.04864

Augmented Bayesian Policy Search

Deterministic policies are often preferred over stochastic ones when implemented on physical systems. They can prevent erratic and harmful behaviors while being easier to implement and interpret. However, in practice, exploration is largely performed by stochastic policies. First-order Bayesian Optimization (BO) methods offer a principled way of performing exploration using deterministic policies. This is done through a learned probabilistic model of the objective function and its gradient. Nonetheless, such approaches treat policy search as a black-box problem, and thus, neglect the reinforcement learning nature of the problem. In this work, we leverage the performance difference lemma to introduce a novel mean function for the probabilistic model. This results in augmenting BO methods with the action-value function. Hence, we call our method Augmented Bayesian Search~(ABS). Interestingly, this new mean function enhances the posterior gradient with the deterministic policy gradient, effectively bridging the gap between BO and policy gradient methods. The resulting algorithm combines the convenience of the direct policy search with the scalability of reinforcement learning. We validate ABS on high-dimensional locomotion problems and demonstrate competitive performance compared to existing direct policy search schemes.

arXiv:2407.02765

Graphon Particle Systems, Part II: Dynamics of Distributed Stochastic Continuum Optimization

We study the distributed optimization problem over a graphon with a continuum of nodes, which is regarded as the limit of the distributed networked optimization as the number of nodes goes to infinity. Each node has a private local cost function. The global cost function, which all nodes cooperatively minimize, is the integral of the local cost functions on the node set. We propose stochastic gradient descent and gradient tracking algorithms over the graphon. We establish a general lemma for the upper bound estimation related to a class of time-varying differential inequalities with negative linear terms, based upon which, we prove that for both kinds of algorithms, the second moments of the nodes' states are uniformly bounded. Especially, for the stochastic gradient tracking algorithm, we transform the convergence analysis into the asymptotic property of coupled nonlinear differential inequalities with time-varying coefficients and develop a decoupling method. For both kinds of algorithms, we show that by choosing the time-varying algorithm gains properly, all nodes' states achieve $\mathcal{L}^{\infty}$-consensus for a connected graphon. Furthermore, if the local cost functions are strongly convex, then all nodes' states converge to the minimizer of the global cost function and the auxiliary states in the stochastic gradient tracking algorithm converge to the gradient value of the global cost function at the minimizer uniformly in mean square.

arXiv:2407.01402

Superconstant Inapproximability of Decision Tree Learning

We consider the task of properly PAC learning decision trees with queries. Recent work of Koch, Strassle, and Tan showed that the strictest version of this task, where the hypothesis tree $T$ is required to be optimally small, is NP-hard. Their work leaves open the question of whether the task remains intractable if $T$ is only required to be close to optimal, say within a factor of 2, rather than exactly optimal. We answer this affirmatively and show that the task indeed remains NP-hard even if $T$ is allowed to be within any constant factor of optimal. More generally, our result allows for a smooth tradeoff between the hardness assumption and the inapproximability factor. As Koch et al.'s techniques do not appear to be amenable to such a strengthening, we first recover their result with a new and simpler proof, which we couple with a new XOR lemma for decision trees. While there is a large body of work on XOR lemmas for decision trees, our setting necessitates parameters that are extremely sharp, and are not known to be attainable by existing XOR lemmas. Our work also carries new implications for the related problem of Decision Tree Minimization.

arXiv:2406.20054

To Word Senses and Beyond: Inducing Concepts with Contextualized Language Models

Polysemy and synonymy are two crucial interrelated facets of lexical ambiguity. While both phenomena are widely documented in lexical resources and have been studied extensively in NLP, leading to dedicated systems, they are often being considered independently in practical problems. While many tasks dealing with polysemy (e.g. Word Sense Disambiguiation or Induction) highlight the role of word's senses, the study of synonymy is rooted in the study of concepts, i.e. meanings shared across the lexicon. In this paper, we introduce Concept Induction, the unsupervised task of learning a soft clustering among words that defines a set of concepts directly from data. This task generalizes Word Sense Induction. We propose a bi-level approach to Concept Induction that leverages both a local lemma-centric view and a global cross-lexicon view to induce concepts. We evaluate the obtained clustering on SemCor's annotated data and obtain good performance (BCubed F1 above 0.60). We find that the local and the global levels are mutually beneficial to induce concepts and also senses in our setting. Finally, we create static embeddings representing our induced concepts and use them on the Word-in-Context task, obtaining competitive performance with the State-of-the-Art.

arXiv:2406.15682

Background results for robust minmax control of linear dynamical systems

The purpose of this note is to summarize the arguments required to derive the results appearing in robust minmax control of linear dynamical systems using a quadratic stage cost. The main result required in robust minmax control is Proposition 20.a. Moreover, the solution to the trust-region problem given in Proposition 15 and Lemma 16 may be of more general interest. This revised (second) version provides the following corrections and extensions of the previous (first) version. 1. The optimal u and w formulas in the original Corollary 13, Proposition 14, Corollary 19, and Proposition 20 have been corrected in this revision. 2. Corollary 13 and Proposition 14 are combined in the revised Proposition 14.a. 3. Corollary 19 and Proposition 20 are combined in the revised Proposition 20.a. 4. The revised Proposition 12.a is a generalization of the previous Proposition 12. 5. Propositions 5.a, 5.b, and 5.c are new in this revision. 6. Figure 1 has been revised to illustrate the revised Proposition 14.a.

arXiv:2406.16249

An Optimal Tightness Bound for the Simulation Lemma

We present a bound for value-prediction error with respect to model misspecification that is tight, including constant factors. This is a direct improvement of the "simulation lemma," a foundational result in reinforcement learning. We demonstrate that existing bounds are quite loose, becoming vacuous for large discount factors, due to the suboptimal treatment of compounding probability errors. By carefully considering this quantity on its own, instead of as a subcomponent of value error, we derive a bound that is sub-linear with respect to transition function misspecification. We then demonstrate broader applicability of this technique, improving a similar bound in the related subfield of hierarchical abstraction.

arXiv:2406.12530

Discrete-Time Conewise Linear Systems with Finitely Many Switches

We investigate discrete-time conewise linear systems (CLS) for which all the solutions exhibit a finite number of switches. By switches, we mean transitions of a solution from one cone to another. Our interest in this class of CLS comes from the optimization-based control of an insulin infusion model for which the fact that solutions switch finitely many times appears to be key to establish the global exponential stability of the origin. The stability analysis of this class of CLS greatly simplifies compared to general CLS as all solutions eventually exhibit linear dynamics. The main challenge is to characterize CLS satisfying this finite number of switches property. We first present general conditions in terms of set intersections for this purpose. To ease the testing of these conditions, we translate them as a non-negativity test of linear forms using Farkas lemma. As a result, the problem reduces to verify the non-negativity of a single solution to an auxiliary linear discrete-time system. Interestingly, this property differs from the classical non-negativity problem, where any solution to a system must remain non-negative (component-wise) for any non-negative initial condition, and thus requires novel tools to test it. We finally illustrate the relevance of the presented results on the optimal insulin infusion problem.

arXiv:2406.12498

Frequency-Domain Data-Driven Predictive Control

In this paper, we propose a data-driven predictive control scheme based on measured frequency-domain data of the plant. This novel scheme complements the well-known data-driven predictive control (DeePC) approach based on time series data. To develop this new frequency-domain data-driven predictive control (FreePC) scheme, we introduce a novel version of Willems' fundamental lemma based on frequency-domain data. By exploiting frequency-domain data, we allow recent direct data-driven (predictive) control methodologies to benefit from the available expertise and techniques for non-parametric frequency-domain identification in academia and industry. We prove that, under appropriate conditions, the new FreePC scheme is equivalent to the corresponding DeePC scheme. The strengths of FreePC are demonstrated in a numerical case study.

arXiv:2406.09563

e-COP : Episodic Constrained Optimization of Policies

In this paper, we present the $\texttt{e-COP}$ algorithm, the first policy optimization algorithm for constrained Reinforcement Learning (RL) in episodic (finite horizon) settings. Such formulations are applicable when there are separate sets of optimization criteria and constraints on a system's behavior. We approach this problem by first establishing a policy difference lemma for the episodic setting, which provides the theoretical foundation for the algorithm. Then, we propose to combine a set of established and novel solution ideas to yield the $\texttt{e-COP}$ algorithm that is easy to implement and numerically stable, and provide a theoretical guarantee on optimality under certain scaling assumptions. Through extensive empirical analysis using benchmarks in the Safety Gym suite, we show that our algorithm has similar or better performance than SoTA (non-episodic) algorithms adapted for the episodic setting. The scalability of the algorithm opens the door to its application in safety-constrained Reinforcement Learning from Human Feedback for Large Language or Diffusion Models.

arXiv:2406.05637

A Generalized Version of Chung's Lemma and its Applications

Chung's lemma is a classical tool for establishing asymptotic convergence rates of (stochastic) optimization methods under strong convexity-type assumptions and appropriate polynomial diminishing step sizes. In this work, we develop a generalized version of Chung's lemma, which provides a simple non-asymptotic convergence framework for a more general family of step size rules. We demonstrate broad applicability of the proposed generalized Chung's lemma by deriving tight non-asymptotic convergence rates for a large variety of stochastic methods. In particular, we obtain partially new non-asymptotic complexity results for stochastic optimization methods, such as stochastic gradient descent and random reshuffling, under a general $(\theta,\mu)$-Polyak-Lojasiewicz (PL) condition and for various step sizes strategies, including polynomial, constant, exponential, and cosine step sizes rules. Notably, as a by-product of our analysis, we observe that exponential step sizes can adapt to the objective function's geometry, achieving the optimal convergence rate without requiring exact knowledge of the underlying landscape. Our results demonstrate that the developed variant of Chung's lemma offers a versatile, systematic, and streamlined approach to establish non-asymptotic convergence rates under general step size rules.

arXiv:2406.05375

LEMMA-RCA: A Large Multi-modal Multi-domain Dataset for Root Cause Analysis

Root cause analysis (RCA) is crucial for enhancing the reliability and performance of complex systems. However, progress in this field has been hindered by the lack of large-scale, open-source datasets tailored for RCA. To bridge this gap, we introduce LEMMA-RCA, a large dataset designed for diverse RCA tasks across multiple domains and modalities. LEMMA-RCA features various real-world fault scenarios from IT and OT operation systems, encompassing microservices, water distribution, and water treatment systems, with hundreds of system entities involved. We evaluate the quality of LEMMA-RCA by testing the performance of eight baseline methods on this dataset under various settings, including offline and online modes as well as single and multiple modalities. Our experimental results demonstrate the high quality of LEMMA-RCA. The dataset is publicly available at https://lemma-rca.github.io/.

arXiv:2406.04130

An overview of systems-theoretic guarantees in data-driven model predictive control

The development of control methods based on data has seen a surge of interest in recent years. When applying data-driven controllers in real-world applications, providing theoretical guarantees for the closed-loop system is of crucial importance to ensure reliable operation. In this review, we provide an overview of data-driven model predictive control (MPC) methods for controlling unknown systems with guarantees on systems-theoretic properties such as stability, robustness, and constraint satisfaction. The considered approaches rely on the Fundamental Lemma from behavioral theory in order to predict input-output trajectories directly from data. We cover various setups, ranging from linear systems and noise-free data to more realistic formulations with noise and nonlinearities, and we provide an overview of different techniques to ensure guarantees for the closed-loop system. Moreover, we discuss avenues for future research that may further improve the theoretical understanding and practical applicability of data-driven MPC.

arXiv:2406.04012

Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians

Variational inference (VI) is a popular approach in Bayesian inference, that looks for the best approximation of the posterior distribution within a parametric family, minimizing a loss that is typically the (reverse) Kullback-Leibler (KL) divergence. Despite its empirical success, the theoretical properties of VI have only received attention recently, and mostly when the parametric family is the one of Gaussians. This work aims to contribute to the theoretical study of VI in the non-Gaussian case by investigating the setting of Mixture of Gaussians with fixed covariance and constant weights. In this view, VI over this specific family can be casted as the minimization of a Mollified relative entropy, i.e. the KL between the convolution (with respect to a Gaussian kernel) of an atomic measure supported on Diracs, and the target distribution. The support of the atomic measure corresponds to the localization of the Gaussian components. Hence, solving variational inference becomes equivalent to optimizing the positions of the Diracs (the particles), which can be done through gradient descent and takes the form of an interacting particle system. We study two sources of error of variational inference in this context when optimizing the mollified relative entropy. The first one is an optimization result, that is a descent lemma establishing that the algorithm decreases the objective at each iteration. The second one is an approximation error, that upper bounds the objective between an optimal finite mixture and the target distribution.

arXiv:2406.01389

RL in Latent MDPs is Tractable: Online Guarantees via Off-Policy Evaluation

In many real-world decision problems there is partially observed, hidden or latent information that remains fixed throughout an interaction. Such decision problems can be modeled as Latent Markov Decision Processes (LMDPs), where a latent variable is selected at the beginning of an interaction and is not disclosed to the agent. In the last decade, there has been significant progress in solving LMDPs under different structural assumptions. However, for general LMDPs, there is no known learning algorithm that provably matches the existing lower bound (Kwon et al., 2021). We introduce the first sample-efficient algorithm for LMDPs without any additional structural assumptions. Our result builds off a new perspective on the role of off-policy evaluation guarantees and coverage coefficients in LMDPs, a perspective, that has been overlooked in the context of exploration in partially observed environments. Specifically, we establish a novel off-policy evaluation lemma and introduce a new coverage coefficient for LMDPs. Then, we show how these can be used to derive near-optimal guarantees of an optimistic exploration algorithm. These results, we believe, can be valuable for a wide range of interactive learning problems beyond LMDPs, and especially, for partially observed environments.

arXiv:2405.15482

An input-output continuous-time version of Willems' lemma

We illustrate a novel version of Willems' lemma for data-based representation of continuous-time systems. The main novelties compared to previous works are two. First, the proposed framework relies only on measured input-output trajectories from the system and no internal (state) information is required. Second, our system representation makes use of exact system trajectories, without resorting to orthogonal bases representations and consequent approximations. We first establish sufficient and necessary conditions for data-based generation of system trajectories in terms of suitable latent variables. Subsequently, we reformulate these conditions using measured input-output data and show how to span the full behavior of the system. Furthermore, we show how to use the developed framework to solve the data-based continuous-time simulation problem.

arXiv:2405.20724공식 코드 · Papers with Code

Learning on Large Graphs using Intersecting Communities

Message Passing Neural Networks (MPNNs) are a staple of graph machine learning. MPNNs iteratively update each node's representation in an input graph by aggregating messages from the node's neighbors, which necessitates a memory complexity of the order of the number of graph edges. This complexity might quickly become prohibitive for large graphs provided they are not very sparse. In this paper, we propose a novel approach to alleviate this problem by approximating the input graph as an intersecting community graph (ICG) -- a combination of intersecting cliques. The key insight is that the number of communities required to approximate a graph does not depend on the graph size. We develop a new constructive version of the Weak Graph Regularity Lemma to efficiently construct an approximating ICG for any input graph. We then devise an efficient graph learning algorithm operating directly on ICG in linear memory and time with respect to the number of nodes (rather than edges). This offers a new and fundamentally different pipeline for learning on very large non-sparse graphs, whose applicability is demonstrated empirically on node classification tasks and spatio-temporal data processing.

arXiv:2405.20237

Training-efficient density quantum machine learning

Quantum machine learning (QML) requires powerful, flexible and efficiently trainable models to be successful in solving challenging problems. We introduce density quantum neural networks, a model family that prepares mixtures of trainable unitaries, with a distributional constraint over coefficients. This framework balances expressivity and efficient trainability, especially on quantum hardware. For expressivity, the Hastings-Campbell Mixing lemma converts benefits from linear combination of unitaries into density models with similar performance guarantees but shallower circuits. For trainability, commuting-generator circuits enable density model construction with efficiently extractable gradients. The framework connects to various facets of QML including post-variational and measurement-based learning. In classical settings, density models naturally integrate the mixture of experts formalism, and offer natural overfitting mitigation. The framework is versatile - we uplift several quantum models into density versions to improve model performance, or trainability, or both. These include Hamming weight-preserving and equivariant models, among others. Extensive numerical experiments validate our findings.

arXiv:2405.16792

Laurel: Unblocking Automated Verification with Large Language Models

Program verifiers such as Dafny automate proofs by outsourcing them to an SMT solver. This automation is not perfect, however, and the solver often requires hints in the form of assertions, creating a burden for the proof engineer. In this paper, we propose Laurel, a tool that alleviates this burden by automatically generating assertions using large language models (LLMs). To improve the success rate of LLMs in this task, we design two domain-specific prompting techniques. First, we help the LLM determine the location of the missing assertion by analyzing the verifier's error message and inserting an assertion placeholder at that location. Second, we provide the LLM with example assertions from the same codebase, which we select based on a new proof similarity metric. We evaluate our techniques on our new benchmark DafnyGym, a dataset of complex lemmas we extracted from three real-world Dafny codebases. Our evaluation shows that Laurel is able to generate over 56.6\% of the required assertions given only a few attempts, making LLMs an affordable tool for unblocking program verifiers without human intervention.

arXiv:2405.13682

Deep Ridgelet Transform and Unified Universality Theorem for Deep and Shallow Joint-Group-Equivariant Machines

We present a constructive universal approximation theorem for learning machines equipped with joint-group-equivariant feature maps, called the joint-equivariant machines, based on the group representation theory. ``Constructive'' here indicates that the distribution of parameters is given in a closed-form expression known as the ridgelet transform. Joint-group-equivariance encompasses a broad class of feature maps that generalize classical group-equivariance. Particularly, fully-connected networks are not group-equivariant but are joint-group-equivariant. Our main theorem also unifies the universal approximation theorems for both shallow and deep networks. Until this study, the universality of deep networks has been shown in a different manner from the universality of shallow networks, but our results discuss them on common ground. Now we can understand the approximation schemes of various learning machines in a unified manner. As applications, we show the constructive universal approximation properties of four examples: depth-$n$ joint-equivariant machine, depth-$n$ fully-connected network, depth-$n$ group-convolutional network, and a new depth-$2$ network with quadratic forms whose universality has not been known.