检索结果-内蒙古大学图书馆

Covariate-dependent Graphical Model Estimation via Neural Networks with Statistical Guarantees

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Lin, Jiahe Zhang, Yikai Michailidis, George Machine Learning Research Morgan Stanley United States Department of Statistics and Data Science UCLA United States

Graphical models are widely used in diverse application domains to model the conditional dependencies amongst a collection of random variables. In this paper, we consider settings where the graph structure is covariate-dependent, and investigate a deep neural network-based approach to estimate it. The method allows for flexible functional dependency on the covariate, and fits the data reasonably well in the absence of a Gaussianity assumption. Theoretical results with PAC guarantees are established for the method, under assumptions commonly used in an Empirical Risk Minimization framework. The performance of the proposed method is evaluated on several synthetic data settings and benchmarked against existing approaches. The method is further illustrated on real datasets involving data from neuroscience and finance, respectively, and produces interpretable results. © 2025, CC BY.

关键词： Graphic methods

Implicit Regularization for Tubal Tensor Factorizations via Gradient Descent

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Karnik, Santhosh Veselovska, Anna Iwen, Mark Krahmer, Felix Department of Mathematics Northeastern University Boston United States Department of Computational Mathematics Science and Engineering Michigan State University East Lansing United States TUM School of Computation Information and Technology Munich Data Science Institute Technical University of Munich Garching Germany Munich Center for Machine Learning Munich Germany Department of Mathematics Michigan State University East Lansing United States

We provide a rigorous analysis of implicit regularization in an overparametrized tensor factorization problem beyond the lazy training regime. For matrix factorization problems, this phenomenon has been studied in a number of works. A particular challenge has been to design universal initialization strategies which provably lead to implicit regularization in gradient-descent methods. At the same time, it has been argued by [7] that more general classes of neural networks can be captured by considering tensor factorizations. However, in the tensor case, implicit regularization has only been rigorously established for gradient flow or in the lazy training regime. In this paper, we prove the first tensor result of its kind for gradient descent rather than gradient flow. We focus on the tubal tensor product and the associated notion of low tubal rank, encouraged by the relevance of this model for image data. We establish that gradient descent in an overparametrized tensor factorization model with a small random initialization exhibits an implicit bias towards solutions of low tubal rank. Our theoretical findings are illustrated in an extensive set of numerical simulations show-casing the dynamics predicted by our theory as well as the crucial role of using a small random initialization. © 2024, CC BY-SA.

关键词： Matrix factorization

Building a data Lake for Power BI in the Cloud: A Review on Utilizing Cloud Storage Services for Large datasets

学校读者我要写书评

暂无评论

Building a Data Lake for Power BI in the Cloud: A Review on ...

Artificial Intelligence in Education and Industry 4.0 (IDICAIEI), DMIHER International Conference on

作者： Viraj Vijay Gurbade Prateek Verma Swapnil Gundewar Vijendra Singh Bramhe Department of Artificial Intelligence and Data Science Faculty of Engineering and Technology Datta Meghe Institute of Higher Education & Research (DU) Wardha Maharashtra India Department of Artificial Intelligence and Machine Learning Faculty of Engineering and Technology Datta Meghe Institute of Higher Education & Research (DU) Wardha Maharashtra India School of Computing Science and Engineering VIT Bhopal University Sehore Madhya Pradesh India

ISBN: (数字)9798331528713

ISBN: (纸本)9798331528720

A data lake is a centralized repository where you may store both structured and unstructured data at any scale. Unlike typical data warehouses, which need data structure for storage, data lakes allow you to store raw data in its natural state. This adaptability makes data lakes suitable for storing large volumes of various data kinds, including photographs, videos, log files, sensor data, and others. When integrating a data lake with Power Business Intelligence (BI), organizations can harness the power of both technologies to gain valuable insights from their data. In the realm of data-driven decision-making, organizations face the challenge of managing and extracting insights from vast and diverse datasets, a task conventional on-premises storage solutions often struggle with due to the exponential growth of data. Thus, the adoption of more scalable and flexible alternatives becomes imperative. Cloud-based data lakes emerge as transformative solutions, offering unparalleled scalability, accessibility, cost-effectiveness, and advanced security protocols. This paper discuss the concept of constructing a cloud-based data lake optimized for Power BI, leveraging leading cloud storage services like Azure data Lake Storage (ADLS) and Amazon S3. By seamlessly integrating Power BI with the cloud data lake, organizations can overcome traditional storage limitations and reap numerous benefits. The practical applications in advanced analytics, customer insights, IoT data management, and data discovery underscore the real-world utility of cloud data lakes. The integration of Power BI with a cloud data lake augments efficient data analysis and self-service analytics potential, offering organizations a comprehensive solution for deriving actionable insights from their data assets.

关键词： Cloud computing Scalability Decision making Buildings Organizations Big data applications Security data mining Business intelligence Self-service

Mitigating multiple descents: A model-agnostic framework for risk monotonization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Patil, Pratik Kuchibhotla, Arun Kumar Wei, Yuting Rinaldo, Alessandro Department of Statistics and Data Science Carnegie Mellon University PittsburghPA15213 United States Machine Learning Department Carnegie Mellon University PittsburghPA15213 United States Department of Statistics and Data Science The Wharton School University of Pennsylvania PhiladelphiaPA19104 United States

Recent empirical and theoretical analyses of several commonly used prediction procedures reveal a peculiar risk behavior in high dimensions, referred to as double/multiple descent, in which the asymptotic risk is a non-monotonic function of the limiting aspect ratio of the number of features or parameters to the sample size. To mitigate this undesirable behavior, we develop a general framework for risk monotonization based on cross-validation that takes as input a generic prediction procedure and returns a modified procedure whose out-of-sample prediction risk is, asymptotically, monotonic in the limiting aspect ratio. As part of our framework, we propose two data-driven methodologies, namely zero- and one-step, that are akin to bagging and boosting, respectively, and show that, under very mild assumptions, they provably achieve monotonic asymptotic risk behavior. Our results are applicable to a broad variety of prediction procedures and loss functions, and do not require a well-specified (parametric) model. We exemplify our framework with concrete analyses of the minimum l2, l1-norm least squares prediction procedures. As one of the ingredients in our analysis, we also derive novel additive and multiplicative forms of oracle risk inequalities for split cross-validation that are of independent interest. Copyright © 2022, The Authors. All rights reserved.

关键词： Forecasting

Nonlinear Regression with Residuals: Causal Estimation with Time-varying Treatments and Covariates

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Bates, Stephen Kennedy, Edward Tibshirani, Robert Ventura, Valérie Wasserman, Larry Departments of EECS and Statistics University of California Berkeley United States Department of Statistics & Data Science Carnegie Mellon University United States Departments of Biomedical Data Science and Statistics Stanford University United States Department of Statistics Data Science and Neuroscience Institute Carnegie Mellon University United States Departments of Statistics & Data Science and of Machine Learning Carnegie Mellon University United States

Standard regression adjustment gives inconsistent estimates of causal effects when there are time-varying treatment effects and time-varying covariates. Loosely speaking, the issue is that some covariates are post-treatment variables because they may be affected by prior treatment status, and regressing out post-treatment variables causes bias. More precisely, the bias is due to certain non-confounding latent variables that create colliders in the causal graph. These latent variables, which we call phantoms, do not harm the identifiability of the causal effect, but they render naive regression estimates inconsistent. Motivated by this, we ask: how can we modify regression methods so that they hold up even in the presence of phantoms? We develop an estimator for this setting based on regression modeling (linear, log-linear, probit and Cox regression), proving that it is consistent for a reasonable causal estimand. In particular, the estimator is a regression model fit with a simple adjustment for collinearity, making it easy to understand and implement with standard regression software. The proposed estimators are instances of the parametric g-formula, extending the regression-with-residuals approach to several canonical nonlinear models. © 2022, CC BY.

关键词： Phantoms

Vehicle Prediction for BUS Identification Output from our Route Testing Real Time Algorithm

学校读者我要写书评

暂无评论

Vehicle Prediction for BUS Identification Output from our Ro...

Sustainable Communication Networks and Application (ICSCNA), International Conference on

作者： M. Misba R. Ramya Joel Dickson L. Sharmila J. Kavitha K. Udayakumar Department of Artificial Intelligence and Machine Learning Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Tamil Nadu India Department of Artificial Intelligence and Data Science St. Joseph's College of Engineering Chennai Bethlehem Institute of Engineering Karungal Tamil Nadu India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Tamil Nadu India

ISBN: (数字)9798331530013

ISBN: (纸本)9798331530020

Traffic generation in Indian cities is of a mixed nature, comprising over a dozen types of vehicles that can be broadly categorized into slow-moving and fast-moving vehicles. This study highlights the challenges associated with the inefficient operation of the existing bus system. GPS and mobile networks form integral components of the current system, enabling data storage and management using PostgreSQL. The system provides robust and reliable data to fulfill requests from other software applications. Operating on cloud platforms, it is compatible with all major web browsers. The proposed algorithm continuously tracks the real-time location of buses and their corresponding WMS (Web Map Service) location, alerting users through marker notifications. This paper also explores measures to evaluate the efficiency and effectiveness of bus services. Recognizing that an inadequate public transport system impedes economic growth and development, this study emphasizes the need to enhance the overall quality and accessibility of public transportation. The findings aim to improve the operational efficiency of bus services, contributing to a more effective and sustainable transport system.

关键词： Urban areas Software algorithms Memory Prediction algorithms Real-time systems Software Maintenance Software reliability Public transportation Testing

Dynamic Channel Allocation Using Reinforcement learning Algorithm for Multiple Input Multiple Output Systems

学校读者我要写书评

暂无评论

Dynamic Channel Allocation Using Reinforcement Learning Algo...

Integrated Circuits and Communication Systems (ICICACS), IEEE International Conference on

作者： Srinivasa Rao Bittla Raami Riadhusin Divyaraj G N S. Aravindh Ahila B Independent Researcher Department of Computers Techniques Engineering College of Technical Engineering The Islamic University Al Diwaniyah Iraq Department of Artificial Intelligence and Machine Learning Nitte Meenakshi Institute of Technology Bengaluru India Department of Mechanical Engineering New Prince Shri Bhavani College of Engineering and Technology chennai India Department of Artificial Intelligence and Data Science Dhanalakshmi Srinivasan College of Engineering Technology Mamallapuram India

ISBN: (数字)9798331508456

ISBN: (纸本)9798331508463

In recent years, Multiple-Input Multiple-Output systems (MIMO) play a crucial role in modern wireless networks by enhancing spectral efficiency and data rates. Traditional static, heuristic-based allocation methods struggled to adapt to real-time variations, handling high-dimensional state spaces, minimizing interference, and ensuring fairness in channel allocation. However, decision-making in dynamic environments requires intelligent and scalable approach that enables real-time decision-making by efficiently learning and adapting to changing network conditions. To overcome these issues. This research proposed Deep Q-Network with Prioritized Experience Replay (DQN-PER) for dynamic channel allocation for MIMO Systems. DQN-PER prioritizes high-value learning experiences and while also improving training efficiency and convergence speed. PER prioritizes important experiences based on Temporal-Difference (TD) error by allowing the model to focus on more informative transitions, which speeds up learning and improves adaptability. Simulation results demonstrate enhanced spectral efficiency, reduced packet loss, and improved Quality of Service (QoS) in dynamic wireless environments. These findings establish DQN-PER as a promising solution for intelligent and adaptive MIMO resource allocation. The proposed method achieved better results in terms of a packet drop rate 0.04 compared to existing method Advantage Actor-Critic (A2C).

关键词： Spectral efficiency Heuristic algorithms Wireless networks Decision making Quality of service Dynamic scheduling Real-time systems Resource management Optimization Convergence

Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liang, Weixin Izzo, Zachary Zhang, Yaohui Lepp, Haley Cao, Hancheng Zhao, Xuandong Chen, Lingjiao Ye, Haotian Liu, Sheng Huang, Zhi McFarland, Daniel A. Zou, James Y. Department of Computer Science Stanford University United States Machine Learning Department NEC Labs America United States Department of Electrical Engineering Stanford University United States Graduate School of Education Stanford University United States Department of Management Science and Engineering Stanford University United States Department of Computer Science UC Santa Barbara United States Department of Biomedical Data Science Stanford University United States Department of Sociology Stanford University United States Graduate School of Business Stanford University United States

We present an approach for estimating the fraction of text in a large corpus which is likely to be substantially modified or produced by a large language model (LLM). Our maximum likelihood model leverages expert-written and AI-generated reference texts to accurately and efficiently examine real-world LLM-use at the corpus level. We apply this approach to a case study of scientific peer review in AI conferences that took place after the release of ChatGPT: ICLR 2024, NeurIPS 2023, CoRL 2023 and EMNLP 2023. Our results suggest that between 6.5% and 16.9% of text submitted as peer reviews to these conferences could have been substantially modified by LLMs, i.e. beyond spell-checking or minor writing updates. The circumstances in which generated text occurs offer insight into user behavior: the estimated fraction of LLM-generated text is higher in reviews which report lower confidence, were submitted close to the deadline, and from reviewers who are less likely to respond to author rebuttals. We also observe corpus-level trends in generated text which may be too subtle to detect at the individual level, and discuss the implications of such trends on peer review. We call for future interdisciplinary work to examine how LLM use is changing our information and knowledge practices. Copyright © 2024, The Authors. All rights reserved.

关键词： Maximum likelihood

A unified framework for bandit multiple testing 21

学校读者我要写书评

暂无评论

A unified framework for bandit multiple testing

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Ziyu Xu Ruodu Wang Aaditya Ramdas Department of Statistics and Data Science Carnegie Mellon University Department of Statistics and Actuarial Science University of Waterloo Canada Department of Statistics and Data Science Machine Learning Department Carnegie Mellon University

ISBN: (纸本)9781713845393

In bandit multiple hypothesis testing, each arm corresponds to a different null hypothesis that we wish to test, and the goal is to design adaptive algorithms that correctly identify large set of interesting arms (true discoveries), while only mistakenly identifying a few uninteresting ones (false discoveries). One common metric in non-bandit multiple testing is the false discovery rate (FDR). We propose a unified, modular framework for bandit FDR control that emphasizes the decoupling of exploration and summarization of evidence. We utilize the powerful martingale-based concept of "e-processes" to ensure FDR control for arbitrary composite nulls, exploration rules and stopping times in generic problem settings. In particular, valid FDR control holds even if the reward distributions of the arms could be dependent, multiple arms may be queried simultaneously, and multiple (cooperating or competing) agents may be querying arms, covering combinatorial semi-bandit type settings as well. Prior work has considered in great detail the setting where each arm's reward distribution is independent and sub-Gaussian, and a single arm is queried at each step. Our framework recovers matching sample complexity guarantees in this special case, and performs comparably or better in practice. For other settings, sample complexities will depend on the finer details of the problem (composite nulls being tested, exploration algorithm, data dependence structure, stopping rule) and we do not explore these; our contribution is to show that the FDR guarantee is clean and entirely agnostic to these details.

关键词：