107 Hits in 11.1 sec

Real-Time Computation at the Edge of Chaos in Recurrent Neural Networks

Nils Bertschinger, Thomas Natschläger
2004 Neural Computation  
We analyze how the type of dynamics (ordered or chaotic) exhibited by randomly connected networks of threshold gates driven by a time varying input signal depends on the parameters describing the distribution  ...  Hence, this result strongly supports conjectures that dynamical systems which are capable of doing complex computational tasks should operate near the edge of chaos, i.e. the transition from ordered to  ...  Acknowledgments We would like to thank Alexander Kaske for stimulating discussions and helpful comments about the manuscript.  ... 
doi:10.1162/089976604323057443 pmid:15165396 fatcat:j6evksffgjaqnhshi7x2sp4qm4

Exploratory State Representation Learning

Astrid Merckling, Nicolas Perrin-Gilbert, Alex Coninx, Stéphane Doncieux
2022 Frontiers in Robotics and AI  
To solve the problems of exploration and SRL in parallel, we propose a new approach called XSRL (eXploratory State Representation Learning).  ...  Not having access to compact and meaningful representations is known to significantly increase the complexity of reinforcement learning (RL).  ...  The implementation details of the whole neural network are displayed in Table 3. γ = { , , FIGURE 1 .  ... 
doi:10.3389/frobt.2022.762051 pmid:35237669 pmcid:PMC8883277 fatcat:7h63vz56dfc25fyamsqofn5u4i

Review and Perspectives of Machine Learning Methods for Wind Turbine Fault Diagnosis

Mingzhu Tang, Qi Zhao, Huawei Wu, Ziming Wang, Caihua Meng, Yifan Wang
2021 Frontiers in Energy Research  
In the past decades, machine learning (ML) has showed a powerful capability in fault detection and diagnosis of WTs, thereby remarkably reducing equipment downtime and minimizing financial losses.  ...  This study provides a comprehensive review of recent studies on ML methods and techniques for WT fault diagnosis.  ...  and radial basis function (RBF) neural network.  ... 
doi:10.3389/fenrg.2021.751066 fatcat:bzniyxsgofh3pltcnrxd6avhbq

Artificial Intellgence – Application in Life Sciences and Beyond. The Upper Rhine Artificial Intelligence Symposium UR-AI 2021 [article]

Karl-Herbert Schäfer
2021 arXiv   pre-print
, Offenburg and Trier, the Baden-Wuerttemberg Cooperative State University Loerrach, the French university network Alsace Tech (comprised of 14 'grandes \'ecoles' in the fields of engineering, architecture  ...  and management) and the University of Applied Sciences and Arts Northwestern Switzerland.  ...  (Charité, Institute of Pathology) for helpful comments and acknowledge the financial support by the Federal Ministry of Education and Research of Germany (BMBF) in the project deep.HEALTH (13FH770IX6).  ... 
arXiv:2112.05657v1 fatcat:wdjgymicyrfybg5zth2dc2i3ni

Artificial Intelligence in Cardiology: Present and Future

Francisco Lopez-Jimenez, Zachi Attia, Adelaide M. Arruda-Olson, Rickey Carter, Panithaya Chareonthaitawee, Hayan Jouni, Suraj Kapa, Amir Lerman, Christina Luong, Jose R. Medina-Inojosa, Peter A. Noseworthy, Patricia A. Pellikka (+5 others)
2020 Mayo Clinic proceedings  
Artificial intelligence (AI) is a nontechnical, popular term that refers to machine learning of various types but most often to deep neural networks. Cardiology is at the forefront of AI in medicine.  ...  of findings, and addressing safety and ethical concerns before final implementation.  ...  Schematic diagram of the development of a convolutional neural network.  ... 
doi:10.1016/j.mayocp.2020.01.038 pmid:32370835 fatcat:2g6tmuagd5ej7mz3jov4lrdj5i

Prospective Learning: Back to the Future [article]

Joshua T. Vogelstein, Timothy Verstynen, Konrad P. Kording, Leyla Isik, John W. Krakauer, Ralph Etienne-Cummings, Elizabeth L. Ogburn, Carey E. Priebe, Randal Burns, Kwame Kutten, James J. Knierim, James B. Potash (+51 others)
2022 arXiv   pre-print
For example, an intelligence may see a set of pictures of objects, along with their names, and learn to name them.  ...  , and engineer intelligences.  ...  On the number of linear regions of deep neural networks.  ... 
arXiv:2201.07372v1 fatcat:6qktqmoffvd4zl37ubd4wo2f2m

A Review of Some Techniques for Inclusion of Domain-Knowledge into Deep Neural Networks [article]

Tirtharaj Dash, Sharad Chitlangia, Aditya Ahuja, Ashwin Srinivasan
2021 arXiv   pre-print
This paper examines the inclusion of domain-knowledge by means of changes to: the input, the loss-function, and the architecture of deep networks.  ...  We present a survey of ways in which existing scientific knowledge are included when constructing models with neural networks.  ...  implemented by neural networks, and set of rules.  ... 
arXiv:2107.10295v4 fatcat:ifkgq3cptbapfamr3dld57uruu

Situational Understanding in the Human and the Machine

Yan Yufik, Raj Malhotra
2021 Frontiers in Systems Neuroscience  
Suggestions for further R&D are motivated by these hypotheses and are centered on the notions of Active Inference and Virtual Associative Networks.  ...  We define situational understanding and the distinctions between understanding and awareness, consider examples of how understanding—or lack of it—manifest in performance, and review hypotheses concerning  ...  Recent literature associates advanced cognitive capabilities in primates and humans with the ability to monitor the significance of multiple goals in parallel, and to switch between the goals (Mansouri  ... 
doi:10.3389/fnsys.2021.786252 pmid:35002643 pmcid:PMC8733725 fatcat:zeamg7i7rbc6rgh5ev7ffamx5a

Distributed Machine Learning in Materials that Couple Sensing, Actuation, Computation and Communication [article]

Dana Hughes, Nikolaus Correll
2016 arXiv   pre-print
, and corresponding training algorithms, to an amorphous network of computing nodes is considered.  ...  Distributed versions of support vector machines, graphical models and mixture models developed in the field of wireless sensor networks are reviewed.  ...  Artifical neural networks are common approaches to perform classification and regression in the literature cited [28, 29, 33, 36, 48, 84, 91, 92, 97, 113, 140, 142, 147, 149, 161, 162] .  ... 
arXiv:1606.03508v1 fatcat:zb5jcvda5jg6zpsv5w47wwr6nq

TeachMyAgent: a Benchmark for Automatic Curriculum Learning in Deep RL [article]

Clément Romac, Rémy Portelas, Katja Hofmann, Pierre-Yves Oudeyer
2021 arXiv   pre-print
We open-source our environments, all studied ACL algorithms (collected from open-source code or re-implemented), and DRL students in a Python package available at  ...  It includes 1) challenge-specific unit-tests using variants of a procedural Box2D bipedal walker environment, and 2) a new procedural Parkour environment combining most ACL challenges, making it ideal  ...  Learning and development in neural networks: the importance of starting small. Cognition, 48(1):71 - 99, 1993. ISSN 0010-0277. Artif. Life, 25(4):352-365, 2019.  ... 
arXiv:2103.09815v2 fatcat:qtsp6ghsgrd47fwgznnqdwrtsu

Molecular Information Technology

Klaus-Peter Zauner
2005 Critical reviews in solid state and materials sciences  
Investigations into qualitatively new concepts of information processing are underway in the areas of reaction-diffusion computing, self-assembly computing, and conformation-based computing.  ...  Progress will depend on both novel computing concepts and innovations in materials. This article reviews current directions in the use of bulk and single molecules for information processing.  ...  Cellular automata and neural networks provide possible starting points for parallel computing schemes more suitable to molecular implementation.  ... 
doi:10.1080/10408430590918387 fatcat:cg62axmewndhpdf2uo22knwvua

Automatic Curation of Court Documents: Anonymizing Personal Data

Diego Garat, Dina Wonsever
2022 Information  
The challenge was to find a good way of injecting specialized knowledge about person names syntax while taking profit of previous capabilities of pre-trained tools.  ...  There are also interventions in the published data with the aim of generating metadata that enable a better experience of querying and navigation.  ...  Cloze-driven Pretraining of Self-attention Networks.  ... 
doi:10.3390/info13010027 fatcat:ubnfktbvcfhtzl7rgndturoar4

The Tsetlin Machine – A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic [article]

Ole-Christoffer Granmo
2021 arXiv   pre-print
Arguably, the Tsetlin Automaton is an even simpler and more versatile learning mechanism, capable of solving the multi-armed bandit problem.  ...  In five benchmarks, the Tsetlin Machine provides competitive accuracy compared with SVMs, Decision Trees, Random Forests, Naive Bayes Classifier, Logistic Regression, and Neural Networks.  ...  Code Availability Source code and datasets for the Tsetlin Machine, available under the MIT Licence, can be found at and  ... 
arXiv:1804.01508v15 fatcat:ssmvlls2xfdjtbiagwxr5vq6hy

Digital Evolution for Ecology Research: A Review

Emily Dolson, Charles Ofria
2021 Frontiers in Ecology and Evolution  
This work has spanned a wide range of topics, including competition, facilitation, parasitism, predation, and macroecological scaling laws.  ...  We argue for the value of further ecological research in digital evolution systems and present some particularly promising directions for further research.  ...  The process of building the neural network from the genome mirrors development in nature (Channon and Damper, 2000).  ... 
doi:10.3389/fevo.2021.750779 fatcat:kcudtxl5yra7hpb2cjvz3zhlba

Evolution of functional specialization in a morphologically homogeneous robot

Joshua Auerbach, Josh C. Bongard
2009 Proceedings of the 11th Annual conference on Genetic and evolutionary computation - GECCO '09  
E. and J. C. Bongard. (2009). How robot morphology and training order affect the learning of multiple behaviors.  ...  Particular focus is given to methods which evolve not only the control policies of manually-designed robots, but instead evolve both the control policy and physical form of the robot.  ...  Due to the technical challenges of implementing a physical simulator, coupled with the lack of access to computing resources capable of comparable parallelization to the CM-5, it took several years before  ... 
doi:10.1145/1569901.1569915 dblp:conf/gecco/AuerbachB09 fatcat:pj7filcoufc7tb23ux3agpra5u
« Previous Showing results 1 — 15 out of 107 results