This workshop invites works from different strands of the AI community that pertain to the design of algorithms, models, and techniques to deal with smart grids and smart buildings.

Important Dates:

• November 13, 2015 - Extended Submission Deadline
• November 23, 2015 - Acceptance Notification
• December 7, 2015 - Camera-Ready Deadline
• February 12, 2016 - Workshop Date

Registration and hotel information:

• Registration: www.aaai.org/Conferences/AAAI/2016/aaai16registration.php
• Hotel: www.aaai.org/Conferences/AAAI/2016/aaai16hotel.php

The proliferation of intelligent devices and the availability of electric monitoring facilities, broadband communication networks, computational intelligence, and customer-driven electricity storage and generation capabilities, have posed the foundations for the next generation power grids and buildings: smart grids and smart buildings.

Three key aspects distinguish the smart grid from the more traditional electric grid: (1) producers and consumers have access to information (e.g., production costs, customers' electricity needs, time distribution of demands); (2) continuous access to information and communication is possible (e.g., producers and consumers can negotiate prices); and (3) energy can be produced not only by power plants, but also by customers (e.g., via renewable sources, which can be intermittent) and stored for later use (or redistributed through the electric grid). In general terms, a smart grid enables the distributed generation and two-directional flow of electricity, within an integrated system.

Smart buildings form an important component of the smart grid, where technology enables buildings to provide common services (e.g., illumination, thermal comfort, air quality, sanitation, etc.) in a sustainable fashion and at low environmental impact.

AI plays a key role in the smart grid and in smart buildings; the infrastructure provides information to support automated decision making on how to autonomously adapt production and consumption of energy, optimize costs, waste, and environmental impact, and ensure safe, secure, and efficient operation. The goal of this workshop is to bring together researchers from diverse areas of AI to explore both established and novel applications of AI techniques to address problems related to the design, implementation, and deployment of smart grids and smart buildings. Topics include, but are not limited to:

• Multi-agent systems in smart grids and smart buildings
• Optimization methods for smart grids and smart buildings
• Machine learning mechanisms for smart grids and smart buildings
• Knowledge-based methods in design of smart grids and smart buildings
• Coordination of intelligent agents in smart grids and smart buildings
• Human-computer interactions within smart grids and smart buildings
• Negotiation and trading strategies in energy markets
• Simulations of energy markets, smart grids, and smart buildings

Participants should submit a paper (maximum 6 pages + 1 page of references), describing their work on one or more of the topics relevant to the workshop. Accepted papers will be presented during the workshop and will be published as AAAI technical reports, which will be made freely available in AAAI's digital library.

Authors are requested to prepare their papers using the AAAI style files: http://www.aaai.org/Publications/Templates/AuthorKit.zip.

All submissions are conducted via the following website: http://www.easychair.org/conferences/?conf=aisgsb2016.

Submissions should include the name(s), affiliations, and email addresses of all authors in the body of the email. We welcome the submission of papers rejected from the AAAI 2016 technical program. The deadline for receipt of submissions is October 23, 2015. Papers received after this date may not be reviewed.

Submissions will be refereed on the basis of technical quality, novelty, significance, and clarity. Each submission will be thoroughly reviewed by at least two program committee members.

For questions about the submission process, contact the workshop co-chairs.

9:00am - 10:00am: Electricity Disaggregation: A Rich Problem for AI Research
Invited talk by Mario Berges

Electricity disaggregation, or non-intrusive load monitoring (NILM), is a set of techniques used to provide estimates of the consumption of individual electrical appliances in a building from measurements of voltage and/or current at select locations in the facility. In this talk I will describe the NILM problem, and provide a brief historical overview of the solutions that have been proposed to date. I will then turn my attention to a class of solutions based on latent variable models that has received significant attention in recent years, and has lead to encouraging results. As an example of these type of approaches I will describe novel a dual-emission factorial hidden Markov model and an efficient inference procedure that a student in my group is currently working on. The talk will conclude with an overview of the open problems in the field and a discussion of promising paths of inquiry.

10:00am - 10:30am: Automatic Label Correction and Appliance Prioritization in Single Household Electricity Disaggregation
Mark Valovage and Maria Gini

10:30am - 11:00am: Break

11:00am - 11:30am: Planning under Uncertainty for Aggregated Electric Vehicle Charging with Renewable Energy Supply
Erwin Walraven and Matthijs T. J. Spaan

11:30am - 12:00pm: An MDP-Based Winning Approach to Autonomous Power Trading: Formalization and Empirical Analysis
Daniel Urieli and Peter Stone

12:00pm - 12:30pm: Proactive Dynamic DCOPs
Khoi Hoang, Ferdinando Fioretto, Ping Hou, Makoto Yokoo, William Yeoh and Roie Zivan

12:30pm - 2:00pm: Lunch (on your own; no sponsored lunch provided)

2:00pm - 2:30pm: Scalable Causal Learning for Predicting Adverse Events in Smart Buildings
Aniruddha Basak, Ole Mengshoel, Stefan Hosein and Rodney Martin

2:30pm - 3:00pm: Learning to REDUCE: A Reduced Electricity Consumption Prediction Ensemble
Saima Aman, Charalampos Chelmis and Viktor Prasanna

3:00pm - 3:30pm: Active Inference and Dynamic Gaussian Bayesian Networks for Battery Optimization in Wireless Sensor Networks
Caner Komurlu and Mustafa Bilgic

3:30pm - 4:00pm: Break

4:00pm - 4:30pm: Cost-Effective Feature Selection and Ordering for Personalized Energy Estimates
Kirstin Early, Stephen Fienberg and Jennifer Mankoff

4:30pm - 5:00pm: Identifying Contributing Factors of Occupant Thermal Discomfort
Aniruddha Basak, Ole Mengshoel, Stefan Hosein, Rodney Martin, Jayasudha Jayakumaran and Ishwari Aghav

• Mario Berges, Carnegie Mellon University (USA)
• Archie Chapman, University of Sydney (Australia)
• Mathijs de Weerdt, Delft University of Technology (The Netherlands)
• Alessandro Farinelli, University of Verona (Italy)
• Maria Fox, King’s College London (UK)
• Chris Kiekintveld, University of Texas El Paso (USA)
• Akshat Kumar, Singapore Management University (Singapore)
• Amnon Meisels, Ben-Gurion University (Israel)
• Chiara Piacentini, King’s College (UK)
• Sarvapali Ramchurn, University of Southampton (UK)
• Paul Scott, Australian National University and NICTA (Australia)
• Sven Seuken, University of Zurich (Switzerland)
• Pradeep Varakantham, Singapore Management University (Singapore)
• Meritxell Vinyals Salgado, CEA (France)
• William Yeoh, New Mexico State University (USA)

Workshop Co-Chairs:

• Enrico Pontelli, New Mexico State University, epontell@cs.nmsu.edu
• Alex Rogers, University of Southampton, acr@ecs.soton.ac.uk
• Sylvie Thiebaux, Australian National University and NICTA, sylvie.thiebaux@anu.edu.au
• Son Cao Tran, New Mexico State University, tson@cs.nmsu.edu