19 Hits in 9.6 sec

SWIRL: A sequential windowed inverse reinforcement learning algorithm for robot tasks with delayed rewards

Sanjay Krishnan, Animesh Garg, Richard Liaw, Brijen Thananjeyan, Lauren Miller, Florian T Pokorny, Ken Goldberg
2018 The international journal of robotics research  
Sequential Windowed Inverse Reinforcement Learning This section describes an algorithm to infer the parameters for the proposed model.  ...  We present Sequential Windowed Inverse Reinforcement Learning (SWIRL), a three-phase algorithm that partitions a complex task into shorter-horizon subtasks based on Switched Linear Dynamical transitions  ...  Sequential Windowed Inverse Reinforcement Learning (SWIRL) is based on a model for sequential tasks that represents them as a sequence of reward functions R seq = [R 1 , ..., R k ] and transition regions  ... 
doi:10.1177/0278364918784350 fatcat:ze2skzbkfbek5fntflror45ibq

Real-time autonomic decision making under uncertain environments for UAV-based search-and-rescue missions

Vidyasagar Sadhu
For this, I propose an actor-critic based Multi-Agent Deep Reinforcement Learning (MADRL) framework where the critic is trained in a centralized manner and the actor is decentralized and is used during  ...  In this dissertation, I propose novel solutions to deal with environmental, multi-agent non-stationarity, partial observability, and communication uncertainties, and present advanced techniques for real-time  ...  ..., µ θ N } is the set of target policies with delayed parameters θ i and r i is the reward given to UAV i.  ... 
doi:10.7282/t3-ms6y-wr38 fatcat:lnh6thuktvartnxdz6wnmtpkea

Strategies for spatial music performance: the practicalities and aesthetics of responsive systems design

Ricky Graham, Brian Bridges
2015 Divergence Press  
Reviewers Selection The re-new call for participation was published March 29, 2013, and is republished below. A total of 358 submissions were made in all categories of the call.  ...  All submissions in the paper and art tracks were subject to a single-blind peer review process by a group of international experts.  ...  When you click on a text bit the screen zooms in and a window pops up with the specific story.  ... 
doi:10.5920/divp.2015.33 fatcat:xbb7altyobafvfiidg2adzn7l4

On Computational Ecosystems in Media Arts

Rui Filipe Antunes
2016 Leonardo: Journal of the International Society for the Arts, Sciences and Technology  
A Computational Ecosystem is a system of agents designed to emulate, in the computer, biological systems where autonomous individuals are organized in a hierarchical food chain and interact by trading  ...  The central claim is focused on how the narratives normally associated with these systems and their functioning are two complementary, but separate entities.  ...  The ability to learn and adapt, and the possibilities for delays between action and reaction are also present in other techniques of reinforcement learning, such as Neural Networks.  ... 
doi:10.1162/leon_a_01296 fatcat:pjlgq5nkdzgkdmdtypcqf26eme


Tobechukwu ONYEKWENA, Chukwuemeka Ozioma Stanislaus ONWUKWE
2021 Zenodo  
In addition, schedule 5 of Health and Safety at Work etc Act 1974 and the Building Regulations make it mandatory for any contractor or client on a building site to provide first aid boxes, shelter and  ...  accommodation for clothing and for taking meals on every site, protective clothing, fencing and provision of security guard etc to ensure  ...  of a common task.  ... 
doi:10.5281/zenodo.6384380 fatcat:f7edcd535rhkpischquilhadhu

Efficient and robust deep learning

Zhiqiang Tang
Deep learning enables automatically discovering useful, multistage, task-specific features from high-dimensional raw data.  ...  (ii) Chapter 5 introduces the quantized coupled U-Nets architecture to boost the efficiency of stacked U-Nets with broad applications to location-sensitive tasks.  ...  A discounting factor γ is used to model the delaying effects of future rewards.  ... 
doi:10.7282/t3-kjxc-hc73 fatcat:wmtgrs5unzdxdp7b27w5352fsq

Mapping Intermediality in Performance

Maria Chatzichristodoulou
2011 Contemporary Theatre Review  
Last but not least we thank the contributors for their willingness to engage in many dialogues and to adjust their writings to the purposes of the book in what, for us, has been a gratifying, because genuinely  ...  International scholars critically analyze and theorize the materiality and performativity, as well as spatial practices of screen media in contributions that engage with today's digital media culture.  ...  Boenisch, originally from Munich, Germany, is a co-director, with Professors Patrice Pavis and Paul Allain, of the European Theatre Research Network (ETRN) at the University of Kent's School of Arts in  ... 
doi:10.1080/10486801.2011.562056 fatcat:kthirs3lrvhotfbmmm5sqbpjte

Always One Bit More, Computing and the Experience of Ambiguity [chapter]

Fun and Software : Exploring Pleasure, Paradox and Pain in Computing  
these problems, decisions and tasks are formalized through symbols and a set of discrete and finite sequential steps.  ...  in the sequential arrangement of algorithms (randomness works within algorithmic procedures).  ... 
doi:10.5040/ fatcat:vv5cvvow7ngwxkvbqpqgqvejzy

Special Issue: Semantic Informational Technologies

Vladimir Fomichov, Anton Železnikar, Matjaž Gams, Jožef Stefan, Drago Torkar, Jožef Stefan, Editorial Board, Juan Carlos, Augusto, Argentina, Costin Badica, Romania (+20 others)
2010 unpublished
rewarding activity.  ...  Theoretical background for presented system is provided by trialogical learning, an approach in the domain of collaborative learning or working, with several similar aspects to existing constructivist  ...  The primary PR anonymisation is done by the Hospital Information System of the University Specialised Hospital for Active Treatment of Endocrinology "Acad. Acknowledgement  ... 

The development and experience of gesture and brainwave interaction in audiovisual performances [article]

Yago Parra Moutinho Stucky De Quay
that technology, collaborators, and transparency have a decisive impact on a performer's actions and interactive media.  ...  Lastly, I explain how universities and companies can improve research and development by together creating a live show.  ...  Yet the perception of a delayed reaction depended on the task. Trigger gestures required faster systems than those responsible for continuous control.  ... 
doi:10.15781/t2dj58m50 fatcat:uhorj4mz5zd3jbcswib3i4uhoy

ACTA TEHNICA CORVINIENSIS Fascicule 3 [July-September]-Bulletin of Engineering Tome VII [2014] ACTA TEHNICA CORVINIENSIS Fascicule 3 [July-September]-Bulletin of Engineering Tome VII [2014] ACTA TEHNICA CORVINIENSIS Fascicule 3 [July-September]-Bulletin of Engineering Tome VII [2014] ACTA TEHNICA CORVINIENSIS Fascicule 3 [July-September]-Bulletin of Engineering Tome VII [2014] Review process & Editorial Policy

Romania Heput, Hungary Imre, J Rudas, Slovakia Velišek, Serbia Mirjana, Vojinović Miloradov, Alessandro Italy, Gasparetto, Portugal João, Paulo Davim, Romania Carmen, Alic (+82 others)
In a very short period it has acquired global presence and scholars from all over the world have taken it with great enthusiasm.  ...  Like our other journals, ACTA TECHNICA CORVINIENSIS -BULLETIN of ENGINEERING will serve as a great resource for researchers and students across the globe.  ...  This can be used for plain and reinforced concrete with lightweight aggregate and reinforced concrete with normal aggregate.  ... 

Aesthetic Animism: Digital Poetry as Ontological Probe

William David, Jhave Johnston
2011 unpublished
Task-use efficiency is (at a general level) synonymous with compartmentalization.  ...  These are algorithmically tractable tasks that were previously performed by hand pinning keyframes to timelines.  ... 

Chaos Bound: Orderly Disorder in Contemporary Literature and Science

N. Katherine Hayles
For example, turbulent flow can be modeled as small swirls within larger swirls, nested in turn within still larger swirls.  ...  So the robots create "synthetic palefaces," who in turn eventually again make robots.  ...  Why does it appear as a pivotal concept for us here and now?  ... 
doi:10.7298/h4zt-m520 fatcat:uz4x23ssynhy3e5lkmktabc574

Strategies for Unbridled Data Dissemination: An Emergency Operations Manual

Nikita Mazurov
It is found that there exist potential de-anonymization attacks against all analyzed file-sharing tools, with potentially more secure filesharing options also seeing less user adoption.  ...  A close reading of a copyright notice is performed in order to examine its internal contradictions.  ...  they serve to reinforce.  ... 
doi:10.25602/gold.00016589 fatcat:3szv2ve46jb3blsoaalxp5rk7m

Quantum improvisation : sonic consciousness and Pauline Oliveros

Sae-Hoo Stan Chung
Van Nort's FILTER (Freely Improvising, Learning, and Transforming Evolutionary Recombination system) selectively listens and simultaneously plays with the musicians.  ...  Pauline Oliveros's Triple Point (with Doug van Nort and Jonas Braasch) consists of three human musicians performing live with a listening and improvising computer agent.  ...  Virginie Magnat, you offer a living exemplar of the ethical performance scholar, one who seeks compassion and contribution at all times. Words cannot describe the gifts you offered.  ... 
doi:10.14288/1.0074421 fatcat:np233pv4zbd2xm3ie3dsfsjlzq
« Previous Showing results 1 — 15 out of 19 results