18 Hits in 9.6 sec

Edge Computing in the Dark: Leveraging Contextual-Combinatorial Bandit and Coded Computing [article]

Chien-Sheng Yang, Ramtin Pedarsani, A. Salman Avestimehr
2021 arXiv   pre-print
By leveraging the coded computing framework in order to tackle failures or stragglers in computation, we formulate this problem using contextual-combinatorial multi-armed bandits (CC-MAB), and aim to maximize  ...  Our goal is to design an efficient edge computing policy in the dark without the knowledge of the context or computation capabilities of each device.  ...  ONLINE CODED EDGE COMPUTING In this section, we introduce a coded computing framework for the edge computing problem, and formulate the problem as a contextual-combinatorial multi-armed bandit (CC-MAB)  ... 
arXiv:1912.09512v2 fatcat:pmmrbuez6jdslanm6jtws7opv4

Recommendations as Treatments

Thorsten Joachims, Ben London, Yi Su, Adith Swaminathan, Lequn Wang
2021 The AI Magazine  
This article explains how these techniques enable unbiased offline evaluation and learning despite biased data, and how they can inform considerations of fairness and equity in recommender systems.  ...  In recent years, a new line of research has taken an interventional view of recommender systems, where recommendations are viewed as actions that the system takes to have a desired effect.  ...  Dark gray shading represents the logging and target policies.  ... 
doi:10.1609/aimag.v42i3.18141 fatcat:hdyi4nadijgp3fpieqojib5pfq

30th Annual Computational Neuroscience Meeting: CNS*2021–Meeting Abstracts

2021 Journal of Computational Neuroscience  
One of the goals of neuroscience is to understand the computational principles that describe the formation of behaviorally relevant signals in the brain, as well as how these computations are realized  ...  Recently, the framework of predictive coding (Sajikumar et al., 2014  ...  In particular, we wanted to explore the effect of the spatial extension, position and sparsity of the MF input , and in the next stage, the projection patterns from the GrCs to the PCs (Josselyn & Tonegawa  ... 
doi:10.1007/s10827-021-00801-9 pmid:34931275 pmcid:PMC8687879 fatcat:evpmmfpaivgpxdqpive5xdgmwu

26th Annual Computational Neuroscience Meeting (CNS*2017): Part 2

Leonid L. Rubchinsky, Sungwoo Ahn, Wouter Klijn, Ben Cumming, Stuart Yates, Vasileios Karakasis, Alexander Peyser, Marmaduke Woodman, Sandra Diaz-Pier, James Deraeve, Eliana Vassena, William Alexander (+473 others)
2017 BMC Neuroscience  
light and dark blue, respectively): detection threshold decreases 9.  ...  standard armed-bandit problem.  ... 
doi:10.1186/s12868-017-0371-2 fatcat:jkqnv4rvfvaj5g44xm5snzdqzy

Reinforcement Learning for Machine Translation: from Simulations to Real-World Applications

Julia Kreutzer
In einem interaktiven Online-Lernszenario erhält ein MT Modell "Bandit"-Feedback (d.h. für nur je eineÜbersetzung) anstelle von Referenzübersetzungen zum Lernen.  ...  In Produktionsumgebungen wird Offline-Lernen allerdings oft dem Online-Lernen vorgezogen.  ...  Our case is related to contextual bandits (Langford and Zhang, 2007; Li et al., 2010) or combinatorial bandits (Dani et al., 2007; Cesa-Bianchi and Lugosi, 2012) in the sense that the parametric models  ... 
doi:10.11588/heidok.00028862 fatcat:jrsiseo4prf4pa3f7nnbp24wkq

Elastic resource allocation mechanisms for the Cloud based on Reinforcement Learning [article]

Alexandros Kontarinis, National Technological University Of Athens, National Technological University Of Athens
Cloud Computing, thanks to its core idea of multiple users sharing the same physical resources, has made possible to scale computing resources up and down, at will, and with minimal friction.  ...  In specific, we examine a range of Reinforcement Learning and Multi-Armed Bandit algorithms, methods, and techniques, for achieving efficient adaptive cloud resource provisioning.  ...  These are Value-Function approximation and Contextual Bandits.  ... 
doi:10.26240/heal.ntua.9611 fatcat:uwv2pmf3ufgfrhtex3qb2hg4ua

ACNP 57th Annual Meeting: Poster Session II

2018 Neuropsychopharmacology  
Increasing evidence suggests a role for systemic and neurological inflammation in the pathophysiology of fear and trauma exposure based psychiatric disorders (Micholpoulos et al., 2017; Haroon et al.,  ...  After one week, mice were tested for avoidance behaviors by open field and light-dark box testing.  ...  ' behavior on a multi-arm bandit task.  ... 
doi:10.1038/s41386-018-0267-6 fatcat:febeq6uwefgdzm65ccmzrmjwjy

Self Exploration of Sensorimotor Spaces in Robots. Self-Exploration of Sensorimotor Spaces in Robots Mathématiques et Informatique Spécialité Informatique L'auto-exploration des espaces sensorimoteurs chez les robots Mathématiques et Informatique Spécialité Informatique

Fabien Benureau, Fabien Benureau, Verena Hafner, Jean-Baptiste Mouret, Manuel Lopes, Fabien Benureau, Verena Hafner, Jean-Baptiste Mouret
In order to allow unrestricted examination and reproduction of all our results, the entire code is made available.  ...  and embodied experience of the real-world.  ...  In other words, when fumbling in the dark, diversity is a powerful tool.  ... 

Quality of Experience: From Assessment to Application (Dagstuhl Seminar 15022) Understanding Complexity in Multiobjective Optimization (Dagstuhl Seminar 15031) Model-driven Algorithms and Architectures for Self-Aware Computing Systems (Dagstuhl Seminar 15041) Artificial and Computational Intelligence in Games: Integration (Dagstuhl Seminar 15051)

Michael Philippsen, Pascal Felber, Michael Scott, J Eliot, Katrien De Moor, Markus Fiedler, Peter Reichl, Martín Varela, Salvatore Greco, Kathrin Klamroth, Joshua Knowles, Günter Rudolph (+13 others)
2015 unpublished
The research field of artificial and computational intelligence in games focuses on the wide variety of advanced computational techniques used to create, enhance, and improve the experiences of humans  ...  In the working group discussions in Dagstuhl, Ralph E. Steuer presented interesting visualization challenges. Carlos Henggeler Antunes and Carlos M.  ...  /simulation/ computer-codes/visit).  ... 

End-to-end deep reinforcement learning in computer systems

Michael Schaarschmidt, Apollo-University Of Cambridge Repository, Apollo-University Of Cambridge Repository, Eiko Yoneki
In this dissertation, I first argue that in order to design and execute deep RL algorithms efficiently, novel software abstractions are required which can accommodate the distinct computational patterns  ...  I argue that progress in applied RL is hindered by a lack of tools for task model design which bridge the gap between systems and algorithms, and also by missing shared standards for evaluation of model  ...  Third, even with tools for automatic graph generation from imperative code, algorithms need to be structured in a way that allows to leverage graph generation to connect computation fragments into end-to-end  ... 
doi:10.17863/cam.51466 fatcat:lw3m42oa7zaktoqvhzwhhci4qm

Robot grasping by exploiting compliance and environmental constraints

Clemens Eppner, Technische Universität Berlin, Technische Universität Berlin, Oliver Brock
Next, we focus on the interactions between hand and environment. We show that contacting the environment can improve success in motion and grasping tasks. We conclude our in [...]  ...  In this thesis, we exploit two general properties to devise grasp planning algorithms: the compliance of robot hands and the stiffness of the environment that surrounds an object.  ...  Thus, a more suitable formalization is the contextual multi-armed bandit problem (Langford and Zhang, 2008) .  ... 
doi:10.14279/depositonce-8896 fatcat:2pncbuw7s5ccvoem746qoogwe4

Abstracts of Working Papers in Economics

2001 Abstracts of Working Papers in Economics  
AB We examine HMO participation and enrollment in the Medicare risk market for the years 1990 to 1995.  ...  We also find empirical support for the existence of complementarities in the joint production of an HMO's commercial and Medicare products. Acemoglu, Daron  ...  Bayesian estimation of the reduced rank regression model without ordering restrictions was presented in Strachan (1998) and follows the classical approach of Anderson (1951) and Johansen (1988).  ... 
doi:10.1017/s0951007900005398 fatcat:vp6a7tbsqvf7xb3gpis5kr2fde

Schloss Dagstuhl - Jahresbericht / Annual Report 2017

Schloss Dagstuhl-Leibniz-Zentrum Für Informatik
The unique scientific atmosphere and the historic building provided not only all the room we needed for our program and the working groups, but also plenty of opportunities for continued discussions and  ...  Finally, we are grateful to and Seda Guerses for taking notes during two of the working groups and to Savvas Zannettou for coordinating the writing of this report and taking notes throughout the seminar  ...  When you feel alone in the dark, shout and you may find that others are there in the tunnel with you.  ... 
doi:10.4230/2199-1995.2017 fatcat:im34ekmmxnbrtlgjk3os6u53pq

Visual Similarity and Representation Learning

Timo Milbich
Thank you Lisa, Noah, Sabine, Uta, Fabrizio, Robert, Pablo and Biagio for all the moments of joy and laughter, tolerating my bad jokes, but most of all for supporting me in difficult moments and dark times  ...  Christoph Schnörr for his interest in my work and for his willingness to review my thesis.  ...  Such a setup is also successfully adopted by contextual bandits [138] 1 .  ... 
doi:10.11588/heidok.00030731 fatcat:zqwaiiphivhdvc7nkpk3t3e7ay

Stampede Theory: Mapping and Treating Dangerous Misinformation at Scale [article]

Philip Gregory Feldman, Maryland Shared Open Access Repository, Wayne ; Lutters
We communicate with each other through computers. Our networks include more than the obvious technologies like VOIP or social media.  ...  In stampedes, social influence becomes the dominant factor, outweighing environmental cues and creating a subjective reality.  ...  The equations described here are encoded and used in the simulation's Java code.  ... 
doi:10.13016/m2z6zo-m1lc fatcat:it6nkrwunrazzcijvcvt2saity
« Previous Showing results 1 — 15 out of 18 results