检索结果-内蒙古大学图书馆

BRAINLM: A FOUNDATION MODEL FOR BRAIN ACTIVITY RECORDINGS 12

学校读者我要写书评

暂无评论

BRAINLM: A FOUNDATION MODEL FOR BRAIN ACTIVITY RECORDINGS

12th International Conference on Learning Representations, ICLR 2024

作者： Caro, Josue Ortega de Fonseca, Antonio H. Rizvi, Syed A. Rosati, Matteo Averill, Christopher Cross, James L. Mittal, Prateek Zappala, Emanuele Dhodapkar, Rahul M. Abdallah, Chadi G. van Dijk, David Wu Tsai Institute United States Interdepartmental Neuroscience Program Yale University United States Baylor College of Medicine United States Department of Computer Science United States Yale School of Medicine United States University of Southern California United States Interdepartmental Program in Computational Biology & Bioinformatics United States Internal Medicine United States Cardiovascular Research Center Yale University United States Department of Mathematics and Statistics Idaho State University United States

We introduce the Brain Language Model (BrainLM), a foundation model for brain activity dynamics trained on 6,700 hours of fMRI recordings. Utilizing self-supervised masked-prediction training, BrainLM demonstrates proficiency in both fine-tuning and zero-shot inference tasks. Fine-tuning allows for the accurate prediction of clinical variables like age, anxiety, and PTSD as well as forecasting of future brain states. Critically, the model generalizes well to entirely new external cohorts not seen during training. In zero-shot inference mode, BrainLM can identify intrinsic functional networks directly from raw fMRI data without any network-based supervision during training. The model also generates interpretable latent representations that reveal relationships between brain activity patterns and cognitive states. Overall, BrainLM offers a versatile and interpretable framework for elucidating the complex spatiotemporal dynamics of human brain activity. It serves as a powerful "lens" through which massive repositories of fMRI data can be analyzed in new ways, enabling more effective interpretation and utilization at scale. The work demonstrates the potential of foundation models to advance computational neuroscience research. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Brain

Cupolets in a Chaotic Neuron Model

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Parker, John E. Short, Kevin M. Department of Mathematics University of Pittsburgh PittsburghPA15260 United States Integrated Applied Mathematics Program Department of Mathematics and Statistics University of New Hampshire DurhamNH03824 United States

This paper reports the first finding of cupolets in a chaotic Hindmarsh-Rose neural model. Cupolets (chaotic, unstable, periodic, orbit-lets) are unstable periodic orbits that have been stabilized through a particular control scheme applying a binary control sequence. We demonstrate different neural dynamics (periodic or chaotic) of the Hindmarsh-Rose model through a bifurcation diagram where the external input current, I, is the bifurcation parameter. We select a region in the chaotic parameter space and provide the results of numerical simulations. In this chosen parameter space, a control scheme is applied when the trajectory intersects with either of the two control planes. The size of the control is determined by a bit in a binary control sequence. The control is either a small microcontrol (0) or a large macrocontrol (1) that adjusts the future dynamics of the trajectory. We report the discovery of many cupolets with corresponding control sequences and comment on the differences with previously reported cupolets in the double scroll system. We provide some examples of the generated cupolets and conclude by discussing potential implications for biological neurons. © 2022, CC BY.

关键词： Bifurcation (mathematics)

CaPS: Collaborative and Private Synthetic Data Generation from Distributed Sources

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Pentyala, Sikha Pereira, Mayana De Cock, Martine School of Engineering and Technology University of Washington Tacoma United States Department of Electrical Engineering Uni-versidade de Brasilia Brazil Department of Applied Mathematics Computer Science and Statistics Ghent University Belgium

Data is the lifeblood of the modern world, forming a fundamental part of AI, decision-making, and research advances. With increase in interest in data, governments have taken important steps towards a regulated data world, drastically impacting data sharing and data usability and resulting in massive amounts of data confined within the walls of organizations. While synthetic data generation (SDG) is an appealing solution to break down these walls and enable data sharing, the main drawback of existing solutions is the assumption of a trusted aggregator for generative model training. Given that many data holders may not want to, or be legally allowed to, entrust a central entity with their raw data, we propose a framework for collaborative and private generation of synthetic tabular data from distributed data holders. Our solution is general, applicable to any marginal-based SDG, and provides input privacy by replacing the trusted aggregator with secure multi-party computation (MPC) protocols and output privacy via differential privacy (DP). We demonstrate the applicability and scalability of our approach for the state-of-the-art select-measure-generate SDG algorithms MWEM+PGM and AIM. Copyright © 2024, The Authors. All rights reserved.

关键词： Decision making

On seeded subgraph-to-subgraph matching: The ssSGM Algorithm and matchability information theory

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Meng, Lingyao Lou, Mengqi Lin, Jianyu Lyzinski, Vince Fishkind, Donniell E. Johns Hopkins University Department of Applied Mathematics and Statistics United States Georgia Institute of Technology School of Industrial and Systems Engineering United States University of Maryland Department of Mathematics College Park United States

The subgraph-subgraph matching problem is, given a pair of graphs and a positive integer K, to find K vertices in the first graph, K vertices in the second graph, and a bijection between them, so as to minimize the number of adjacency disagreements across the bijection;it is "seeded" if some of this bijection is fixed. The problem is intractable, and we present the ssSGM algorithm, which uses Frank-Wolfe methodology to efficiently find an approximate solution. Then, in the context of a generalized correlated random Bernoulli graph model, in which the pair of graphs naturally have a core of K matched pairs of vertices, we provide and prove mild conditions for the subgraph-subgraph matching problem solution to almost always be the correct K matched pairs of *** Codes 05C60, 05C80, 90C35 © 2023, CC BY.

关键词： Information theory

Dual Adversarial Alignment for Realistic Support-Query Shift Few-shot Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jiang, Siyang Fang, Rui Chen, Hsi-Wen Ding, Wei Chen, Ming-Syan School of Mathematics and Statistics Huizhou University Huizhou China Department of Electrical and Engineering National Taiwan University TW Taipei Taiwan

Support-query shift few-shot learning aims to classify unseen examples (query set) to labeled data (support set) based on the learned embedding in a low-dimensional space under a distribution shift between the support set and the query set. However, in real-world scenarios the shifts are usually unknown and varied, making it difficult to estimate in advance. Therefore, in this paper, we propose a novel but more difficult challenge, RSQS, focusing on Realistic Support-Query Shift few-shot learning. The key feature of RSQS is that the individual samples in a meta-task are subjected to multiple distribution shifts in each meta-task. In addition, we propose a unified adversarial feature alignment method called DUal adversarial ALignment framework (DuaL) to relieve RSQS from two aspects, i.e., inter-domain bias and intra-domain variance. On the one hand, for the inter-domain bias, we corrupt the original data in advance and use the synthesized perturbed inputs to train the repairer network by minimizing distance in the feature level. On the other hand, for intra-domain variance, we proposed a generator network to synthesize hard, i.e., less similar, examples from the support set in a self-supervised manner and introduce regularized optimal transportation to derive a smooth optimal transportation plan. Lastly, a benchmark of RSQS is built with several state-of-the-art baselines among three datasets (CIFAR100, mini-ImageNet, and Tiered-Imagenet). Experiment results show that DuaLsignificantly outperforms the state-of-the-art methods in our benchmark. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

AutoRank: MCDA Based Rank Personalization for LoRA-Enabled Distributed Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Shuaijun Tavallaie, Omid Nazimi, Niousha Chen, Xin Zomaya, Albert Y. School of Computer Science The University of Sydney Australia Department of Engineering Science University of Oxford United Kingdom School of Mathematics and Statistics Central South University China

As data volumes expand rapidly, distributed machine learning has become essential for addressing the growing computational demands of modern AI systems. However, training models in distributed environments is challenging with participants hold skew, Non-Independent-Identically distributed (Non-IID) data. Low-Rank Adaptation (LoRA) offers a promising solution to this problem by personalizing low-rank updates rather than optimizing the entire model, LoRA-enabled distributed learning minimizes computational and maximize personalization for each participant. Enabling more robust and efficient training in distributed learning settings, especially in large-scale, heterogeneous systems. Despite the strengths of current state-of-the-art methods, they often require manual configuration of the initial rank, which is increasingly impractical as the number of participants grows. This manual tuning is not only time-consuming but also prone to suboptimal configurations. To address this limitation, we propose AutoRank, an adaptive rank-setting algorithm inspired by the bias-variance trade-off. AutoRank leverages the MCDA method TOPSIS to dynamically assign local ranks based on the complexity of each participant’s data. By evaluating data distribution and complexity through our proposed data complexity metrics, AutoRank provides fine-grained adjustments to the rank of each participant’s local LoRA model. This adaptive approach effectively mitigates the challenges of double-imbalanced, non-IID data. Experimental results demonstrate that AutoRank significantly reduces computational overhead, enhances model performance, and accelerates convergence in highly heterogeneous federated learning environments. Through its strong adaptability, AutoRank offers a scalable and flexible solution for distributed machine learning. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

Multi-objective Frequency Conversion Sinusoidal Chaotic Neural Network

学校读者我要写书评

暂无评论

Multi-objective Frequency Conversion Sinusoidal Chaotic Neur...

第43届中国控制会议

作者： Zhiqiang Hu Xiaodong Zhao Xiaoqian Li Lei Wang Zhonghuan Yang Liwei Fei School of Mathematics and Statistics Taishan University School of Automation Beijing Information Science and Technology University Department of Information Engineering Shandong Vocational Institute of Fashion Technology

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

For the defect of Frequency Conversion Sinusoidal Chaotic Neural Network(FCSCNN) cannot solve the multi-objective optimization *** analyzing the action mechanism and design principle of the existing multi-objective intelligent optimization algorithm,this paper designs a multi-objective optimization algorithm model(MOFCSCNN) suitable for FCSCNN from three aspects,namely,the selection and update of global optimal solution,extraction of non-dominant solution and management of *** experiments on ZDT series benchmark functions,the feasibility and effectiveness of the algorithm are *** order to further test the performance of the algorithm,it is compared with several commonly used multi-objective optimization *** to the simulation results,MOFCSCNN algorithm can realize the optimization of multi-objective problems,and shows better performance in convergence and optimization speed than several existing algorithms.

关键词： chaotic neural network multi-objective intelligent optimization non-dominant solution archive

From spacetime effective media to cascaded frequency conversion

学校读者我要写书评

暂无评论

From spacetime effective media to cascaded frequency convers...

URSI Atlantic Radio Science Conference (URSI AT-RASC)

作者： Victor Pacheco-Peña Nader Engheta School of Mathematics Statistics and Physics Newcastle University Newcastle Upon Tyne United Kingdom Department of Electrical and Systems Engineering University of Pennsylvania Philadelphia PA USA

ISBN: (数字)9789463968102

ISBN: (纸本)9798350360257

Wave propagation in materials modulated in space and time has become a new paradigm for a full control of wavematter interactions. Here, we will discuss our recent efforts in the field using time interfaces (i.e., a rapid change in time of the electromagnetic parameters of the medium where a wave travels). Applications include the use of time interfaces to enable space-time effective media, as well as frequency conversion in dielectric slab waveguides (single and cascaded waveguides) where the cladding material is time varying while the core does not change with time.

关键词： Waveguide transitions Claddings Propagation Waveguide theory Frequency conversion Nonhomogeneous media Dielectrics

Interpretable Machine Learning Models Can Outperform Statistical Models in Solder Joint Reliability

学校读者我要写书评

暂无评论

Interpretable Machine Learning Models Can Outperform Statist...

Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITHERM)

作者： Qais Qasaimeh Haoran Li Sa’d Hamasha John Evans Jia Liu Department of Industrial and Systems Engineering Auburn University AL Department of Mathematics and Statistics Auburn University AL The Center for Advanced Vehicle and Extreme Environmental Electronics (CAVE3)

ISBN: (数字)9798350364330

ISBN: (纸本)9798350364347

Weibull regression and the Cox proportional hazard model (Cox-PHM) are commonly employed for analyzing the solder joint reliability due to their simplicity and meaningful results. Recently, various machine learning (ML) models have been applied and shown similar or better performance than conventional approaches, yet they do not incur widespread adoption in reliability modeling because of their limited transparency and interpretability. In this research, we compare the predictive capabilities of Weibull regression and Cox-PHM with ML techniques (specifically, Random Survival Forests [RSF] and Gradient Boosting [GB]) using experimental data from an accelerated thermal cycling reliability test and calculating the concordance index (c-index) as an evaluation metric to assess the effectiveness of a model in predicting the sequence of failure times in the tests. The findings with our dataset showed that ML-based models have a similar performance with the traditional Weibull regression (c-index = 0.808), and GB can even outperform with a high c-index value of 0.848. The superior performance of ML models can be attributed to their capacity to capture nonlinearities and intricate relationships within the data. Additionally, we utilized feature importance scores to understand which factors/variables significantly influence the model's prediction. Such interpretable ML methods can provide clear insights into how models make predictions, thereby enhancing trust and the adoption of innovative ML techniques in electronics manufacturing.

关键词： Measurement Forests Thermomechanical processes Life estimation Predictive models Data models Hazards Reliability Indexes Soldering