2,984 Hits in 4.4 sec

Multi-Advisor Reinforcement Learning [article]

Romain Laroche and Mehdi Fatemi and Joshua Romoff and Harm van Seijen
2017 arXiv   pre-print
We consider tackling a single-agent RL problem by distributing it to n learners. These learners, called advisors, endeavour to solve the problem from a different focus.  ...  advisors disagree, and the agnostic planning is inefficient around danger zones.  ...  MULTI-ADVISOR REINFORCEMENT LEARNING Markov Decision Process -The Reinforcement Learning (RL) framework is formalised as a Markov Decision Process (MDP).  ... 
arXiv:1704.00756v2 fatcat:53ndhvnfgvez5ecvts47ejnpse

The Mythos of Model Interpretability [article]

Zachary C. Lipton
2017 arXiv   pre-print
Supervised machine learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world?  ...  Throughout, we discuss the feasibility and desirability of different notions, and question the oft-made assertions that linear models are interpretable and that deep neural networks are not.  ...  However, like supervised learning, reinforcement learning relies on a well-defined scalar objective.  ... 
arXiv:1606.03490v3 fatcat:5yptgailtveo5nxywjp7ue5uoi

Pragmatic navigation: reactivity, heuristics, and search

Susan L. Epstein
1998 Artificial Intelligence  
Additional contributions of this paper include a FORR-based, pragmatic, cognitively plausible approach to navigation with learned heuristic approximations that describe two-dimensional territory and travel  ...  FORR categorizes methods as reactive, heuristic, or situationbased, and addresses problem solving with one category of methods at a time.  ...  Acknowledgments David Sullivan's preliminary work on Tonka convinced me that a FORR-based version of Ariadne could be a success.  ... 
doi:10.1016/s0004-3702(97)00083-0 fatcat:4p6gwbdrz5bbbiaajl6nlzdljm

An Overview of Machine Learning-Based Techniques for Solving Optimization Problems in Communications and Signal Processing

Hayssam Dahrouj, Rawan Alghamdi, Hibatallah Alwazani, Sarah Bahanshal, Alaa Alameer Ahmad, Alice Faisal, Rahaf Shalabi, Reem Alhadrami, Abdulhamit Subasi, Malak Alnory, Omar Kittaneh, Jeff S. Shamma
2021 IEEE Access  
To address the offloading decision problem, the authors in [84] use deep reinforcement learning which solves the problem with a fast convergence speed.  ...  c: Deep Reinforcement based Online Offloading Algorithm In [84] , the authors propose solving the MIP, non-convex problem in (45) by devising a deep reinforcement learning online offloading algorithm  ... 
doi:10.1109/access.2021.3079639 fatcat:naklpjg5lbaarb7izvkg2wsynu

Computing Graph Neural Networks: A Survey from Algorithms to Accelerators [article]

Sergi Abadal, Akshay Jain, Robert Guirado, Jorge López-Alonso, Eduard Alarcón
2021 arXiv   pre-print
Graph Neural Networks (GNNs) have exploded onto the machine learning scene in recent years owing to their capability to model and learn from graph-structured data.  ...  This includes a brief tutorial on the GNN fundamentals, an overview of the evolution of the field in the last decade, and a summary of operations carried out in the multiple phases of different GNN algorithm  ...  Deep Learning on Graphs: A Survey [181] (2020) • Provides a discussion on graph versions of recurrent and convolutional networks, autoencoders, reinforcement-learning and adversarial methods • Presents  ... 
arXiv:2010.00130v3 fatcat:u5bcmjodcfdh7pew4nssjemdba

Data Science Technologies in Economics and Finance: A Gentle Walk-In [chapter]

Luca Barbaglia, Sergio Consoli, Sebastiano Manzan, Diego Reforgiato Recupero, Michaela Saisana, Luca Tiozzo Pezzoli
2021 Data Science for Economics and Finance  
In economics and finance, in particular, tapping into these data brings research and business closer together, as data generated in ordinary economic activity can be used towards effective and personalized  ...  It also outlines some of the common issues in economic modeling with data science technologies and surveys the relevant big data management and analytics solutions, motivating the use of data science methods  ...  Next to deep learning, reinforcement learning has gained popularity in recent years: it is based on a paradigm of learning by trial and error, solely from rewards or punishments.  ... 
doi:10.1007/978-3-030-66891-4_1 fatcat:yfs57yvdzbaddmpjw5qqn2mdu4

Toward an ideal trainer

Susan L. Epstein
1994 Machine Learning  
The results argue for a broad variety of training experience with play at many levels. This variety may either be inherent in the game or introduced deliberately into the training.  ...  Lesson and practice training, a blend of expert guidance and knowledge-based, self-directed elaboration, is shown to be particularly effective for learning during competition.  ...  Experiments with a simple pattern-learner and reinforcement training for several games on a three-by-three grid have indicated that priming may slow learning (Painter, 1993) .  ... 
doi:10.1007/bf00993346 fatcat:3nrqgcbnmzcxpjqwphdvlhmkze

Edge on Wheels with OMNIBUS Networking in 6G Technology

Mustafa Ergen, Feride Inan, Onur Ergen, Ibraheem Shayea, Mehmet Fatih Tuysuz, Azizul Azizan, Nazim Kemal Ure, Maziar Nekovee
2020 IEEE Access  
His main research interests are applications of deep learning and deep reinforcement learning for autonomous systems, large scale optimization, and development of high-performance guidance navigation and  ...  (i) Ensuring consensus among multiple cars working towards a common goal. For instance, when all cars involved are solving one optimisation problem together, yet with different data set partitions.  ... 
doi:10.1109/access.2020.3038233 fatcat:iapvre5oknb6blnpd7sx3ka2cq

Lessons Learned from a Decade of Providing Interactive, On-Demand High Performance Computing to Scientists and Engineers [chapter]

Julia Mullen, Albert Reuther, William Arcand, Bill Bergeron, David Bestor, Chansup Byun, Vijay Gadepally, Michael Houle, Matthew Hubbell, Michael Jones, Anna Klein, Peter Michaleas (+6 others)
2018 Lecture Notes in Computer Science  
For decades, the use of HPC systems was limited to those in the physical sciences who had mastered their domain in conjunction with a deep understanding of HPC architectures and algorithms.  ...  For over a decade, MIT Lincoln Laboratory has been supporting interactive, on-demand high performance computing by seamlessly integrating familiar high productivity tools to provide users with an increased  ...  These research programs, with applications requiring massive computational effort, prepared students who had the time, inclination and mandate from their advisors to learn how to program and exploit the  ... 
doi:10.1007/978-3-030-02465-9_47 fatcat:ylvlmv36wzfd5c5drqk3vul4hi

IEEE INFOCOM 2020 Workshops: IEEE INFOCOM 2020 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) - Program

2020 IEEE INFOCOM 2020 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS)  
Balancing Data Freshness and Distortion in Real-time Status Updating with Lossy Compression  ...  Protocol A Data Forwarding Mechanism based on Deep Reinforcement Learning for Deterministic Networks Yuhong Li (Beijing University of Posts and Telecommunications, China); Peng Zhang, Yingchao Zhou  ...  He is routinely invited to serve as an advisor to key stakeholder in cellular network eco-system and as a speaker and a panelist on international industrial fora and academic conferences on this topic.  ... 
doi:10.1109/infocomwkshps50562.2020.9163063 fatcat:2cd7o22amjea7goauzo7gm2k7y

Application of machine learning in ophthalmic imaging modalities

Yan Tong, Wei Lu, Yue Yu, Yin Shen
2020 Eye and Vision  
In clinical ophthalmology, a variety of image-related diagnostic techniques have begun to offer unprecedented insights into eye diseases based on morphological datasets with millions of data points.  ...  Machine learning (ML) is an important branch in the field of AI.  ...  , Iterative self-organizing data, fuzzy C-means systems Reinforcement learning Q-learning, Temporal difference learning, State-Action-Reward-State-Action, Teaching-Box systems, Maja systems Deep learning  ... 
doi:10.1186/s40662-020-00183-6 pmid:32322599 pmcid:PMC7160952 fatcat:nwtxlxbwdfdljnupbodgl4v57m

IPDPS 2021 PhD Forum Welcome and Abstracts

2021 2021 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)  
This year, it is being held virtually and will conduct a live session on Monday, May 17th with a lightning round of presentations by the 14 poster authors and a panel to discuss career paths with the PhD  ...  It has grown from the traditional poster presentations by students working toward a PhD in broadly defined areas related to parallel and distributed processing to a broader, enhanced program to include  ...  These circuits are then fed to our deep learning model to infer corresponding fidelities.  ... 
doi:10.1109/ipdpsw52791.2021.00160 fatcat:c5srphikvba6pklpxsyx6lfmhy

Artificial intelligence approaches and mechanisms for big data analytics: a systematic study

Amir Masoud Rahmani, Elham Azhir, Saqib Ali, Mokhtar Mohammadi, Omed Hassan Ahmed, Marwan Yassin Ghafour, Sarkar Hasan Ahmed, Mehdi Hosseinzadeh
2021 PeerJ Computer Science  
The exploration of such huge quantities of data needs more efficient methods with high analysis accuracy.  ...  A number of articles are investigated within each category.  ...  Vu et al. (2020) used a deep learning method to capture the association between the data distribution and the quality of partitioning methods.  ... 
doi:10.7717/peerj-cs.488 pmid:33954253 pmcid:PMC8053021 fatcat:mf6hdb5dqzepnapxyzeacojauu

Representation Learning: A Statistical Perspective [article]

Jianwen Xie, Ruiqi Gao, Erik Nijkamp, Song-Chun Zhu, Ying Nian Wu
2019 arXiv   pre-print
While the origin of learning representations can be traced back to factor analysis and multidimensional scaling in statistics, it has become a central theme in deep learning with important applications  ...  In this article, we review recent advances in learning representations from a statistical perspective.  ...  Learning the generator model jointly with complementary models In modern deep learning literature, the generator model is usually learned jointly with a complementary model, and the learning is not based  ... 
arXiv:1911.11374v1 fatcat:uo47fvw4xndnhm35kr2vjrolpi

Guest Editorial Introduction to the Special Issue on Unmanned Aircraft System Traffic Management

Yan Wan, Ella Atkins, Dengfeng Sun, Kyriakos G. Vamvoudakis, Konstadinos G. Goulias
2021 IEEE transactions on intelligent transportation systems (Print)  
An agile deep reinforcement learning with experience replay model solves the formulated problem concerning the contextual constraints for the UAV-BS navigation.  ...  In [A10] , the article "Data freshness and energy-efficient UAV navigation optimization: a deep reinforcement learning approach," by Abedin et al., designed a navigation policy for multiple UASs where  ... 
doi:10.1109/tits.2021.3101543 fatcat:vvbbvq7emzcjvgje6jceatfty4
« Previous Showing results 1 — 15 out of 2,984 results