Investigation of learning strategies for the SPOT broker in power TAC

Moinul Morshed Porag Chowdhury; Russell Y. Folk; Ferdinando Fioretto; Christopher Kiekintveld; William Yeoh

doi:10.1007/978-3-319-54229-4_7

Investigation of learning strategies for the SPOT broker in power TAC

Moinul Morshed Porag Chowdhury, Russell Y. Folk, Ferdinando Fioretto, Christopher Kiekintveld, William Yeoh

Research output: Chapter in Book/Entry/Poem › Conference contribution

6 Scopus citations

Abstract

The Power TAC simulation emphasizes the strategic problems that broker agents face in managing the economics of a smart grid. The brokers must make trades in multiple markets and, to be successful, brokers must make many good predictions about future supply, demand, and prices in the wholesale and tariff markets. In this paper, we investigate the feasibility of using learning strategies to improve the performance of our broker, SPOT. Specifically, we investigate the use of decision trees and neural networks to predict the clearing price in the wholesale market and the use of reinforcement learning to learn good strategies for pricing our tariffs in the tariff market. Our preliminary results show that our learning strategies are promising ways to improve the performance of the agent for future competitions.

Original language	English (US)
Title of host publication	Agent-Mediated Electronic Commerce
Subtitle of host publication	Designing Trading Strategies and Mechanisms for Electronic Markets - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016
Editors	Ioannis A. Vetsikas, Esther David, Sofia Ceppi, Chen Hajaj, Valentin Robu
Publisher	Springer Verlag
Pages	96-111
Number of pages	16
ISBN (Print)	9783319542287
DOIs	https://doi.org/10.1007/978-3-319-54229-4_7
State	Published - 2017
Externally published	Yes
Event	International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016 - New York, United States Duration: Jul 10 2016 → Jul 10 2016

Publication series

Name	Lecture Notes in Business Information Processing
Volume	271
ISSN (Print)	1865-1348

Conference

Conference	International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016
Country/Territory	United States
City	New York
Period	7/10/16 → 7/10/16

Keywords

Artificial Intelligence
Game theory
Machine learning
Multi agent system
Smart grid

ASJC Scopus subject areas

Management Information Systems
Control and Systems Engineering
Business and International Management
Information Systems
Modeling and Simulation
Information Systems and Management

Access to Document

10.1007/978-3-319-54229-4_7

Cite this

Chowdhury, M. M. P., Folk, R. Y., Fioretto, F., Kiekintveld, C., & Yeoh, W. (2017). Investigation of learning strategies for the SPOT broker in power TAC. In I. A. Vetsikas, E. David, S. Ceppi, C. Hajaj, & V. Robu (Eds.), Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016 (pp. 96-111). (Lecture Notes in Business Information Processing; Vol. 271). Springer Verlag. https://doi.org/10.1007/978-3-319-54229-4_7

Investigation of learning strategies for the SPOT broker in power TAC. / Chowdhury, Moinul Morshed Porag; Folk, Russell Y.; Fioretto, Ferdinando et al.
Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016. ed. / Ioannis A. Vetsikas; Esther David; Sofia Ceppi; Chen Hajaj; Valentin Robu. Springer Verlag, 2017. p. 96-111 (Lecture Notes in Business Information Processing; Vol. 271).

Research output: Chapter in Book/Entry/Poem › Conference contribution

Chowdhury, MMP, Folk, RY, Fioretto, F, Kiekintveld, C & Yeoh, W 2017, Investigation of learning strategies for the SPOT broker in power TAC. in IA Vetsikas, E David, S Ceppi, C Hajaj & V Robu (eds), Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016. Lecture Notes in Business Information Processing, vol. 271, Springer Verlag, pp. 96-111, International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016, New York, United States, 7/10/16. https://doi.org/10.1007/978-3-319-54229-4_7

Chowdhury MMP, Folk RY, Fioretto F, Kiekintveld C, Yeoh W. Investigation of learning strategies for the SPOT broker in power TAC. In Vetsikas IA, David E, Ceppi S, Hajaj C, Robu V, editors, Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016. Springer Verlag. 2017. p. 96-111. (Lecture Notes in Business Information Processing). doi: 10.1007/978-3-319-54229-4_7

Chowdhury, Moinul Morshed Porag ; Folk, Russell Y. ; Fioretto, Ferdinando et al. / Investigation of learning strategies for the SPOT broker in power TAC. Agent-Mediated Electronic Commerce: Designing Trading Strategies and Mechanisms for Electronic Markets - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016. editor / Ioannis A. Vetsikas ; Esther David ; Sofia Ceppi ; Chen Hajaj ; Valentin Robu. Springer Verlag, 2017. pp. 96-111 (Lecture Notes in Business Information Processing).

@inproceedings{f2893573a34d472ba974fb49a1eb7aea,

title = "Investigation of learning strategies for the SPOT broker in power TAC",

abstract = "The Power TAC simulation emphasizes the strategic problems that broker agents face in managing the economics of a smart grid. The brokers must make trades in multiple markets and, to be successful, brokers must make many good predictions about future supply, demand, and prices in the wholesale and tariff markets. In this paper, we investigate the feasibility of using learning strategies to improve the performance of our broker, SPOT. Specifically, we investigate the use of decision trees and neural networks to predict the clearing price in the wholesale market and the use of reinforcement learning to learn good strategies for pricing our tariffs in the tariff market. Our preliminary results show that our learning strategies are promising ways to improve the performance of the agent for future competitions.",

keywords = "Artificial Intelligence, Game theory, Machine learning, Multi agent system, Smart grid",

author = "Chowdhury, {Moinul Morshed Porag} and Folk, {Russell Y.} and Ferdinando Fioretto and Christopher Kiekintveld and William Yeoh",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016 ; Conference date: 10-07-2016 Through 10-07-2016",

year = "2017",

doi = "10.1007/978-3-319-54229-4_7",

language = "English (US)",

isbn = "9783319542287",

series = "Lecture Notes in Business Information Processing",

publisher = "Springer Verlag",

pages = "96--111",

editor = "Vetsikas, {Ioannis A.} and Esther David and Sofia Ceppi and Chen Hajaj and Valentin Robu",

booktitle = "Agent-Mediated Electronic Commerce",

}

TY - GEN

T1 - Investigation of learning strategies for the SPOT broker in power TAC

AU - Chowdhury, Moinul Morshed Porag

AU - Folk, Russell Y.

AU - Fioretto, Ferdinando

AU - Kiekintveld, Christopher

AU - Yeoh, William

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - The Power TAC simulation emphasizes the strategic problems that broker agents face in managing the economics of a smart grid. The brokers must make trades in multiple markets and, to be successful, brokers must make many good predictions about future supply, demand, and prices in the wholesale and tariff markets. In this paper, we investigate the feasibility of using learning strategies to improve the performance of our broker, SPOT. Specifically, we investigate the use of decision trees and neural networks to predict the clearing price in the wholesale market and the use of reinforcement learning to learn good strategies for pricing our tariffs in the tariff market. Our preliminary results show that our learning strategies are promising ways to improve the performance of the agent for future competitions.

AB - The Power TAC simulation emphasizes the strategic problems that broker agents face in managing the economics of a smart grid. The brokers must make trades in multiple markets and, to be successful, brokers must make many good predictions about future supply, demand, and prices in the wholesale and tariff markets. In this paper, we investigate the feasibility of using learning strategies to improve the performance of our broker, SPOT. Specifically, we investigate the use of decision trees and neural networks to predict the clearing price in the wholesale market and the use of reinforcement learning to learn good strategies for pricing our tariffs in the tariff market. Our preliminary results show that our learning strategies are promising ways to improve the performance of the agent for future competitions.

KW - Artificial Intelligence

KW - Game theory

KW - Machine learning

KW - Multi agent system

KW - Smart grid

UR - http://www.scopus.com/inward/record.url?scp=85014164524&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85014164524&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-54229-4_7

DO - 10.1007/978-3-319-54229-4_7

M3 - Conference contribution

AN - SCOPUS:85014164524

SN - 9783319542287

T3 - Lecture Notes in Business Information Processing

SP - 96

EP - 111

BT - Agent-Mediated Electronic Commerce

A2 - Vetsikas, Ioannis A.

A2 - David, Esther

A2 - Ceppi, Sofia

A2 - Hajaj, Chen

A2 - Robu, Valentin

PB - Springer Verlag

T2 - International Workshop on Agent-Mediated Electronic Commerce and Trading Agent Design and Analysis, AMEC/TADA 2016

Y2 - 10 July 2016 through 10 July 2016

ER -

Investigation of learning strategies for the SPOT broker in power TAC

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this