AAAI Spring Symposium Series 2018

Symposium on Integrating
Representation, Reasoning, Learning, and Execution
for Goal Directed Autonomy

March 26th – 28th, 2018

Stanford University


Recent advances in AI and robotics have led to a resurgence of interest in the objective of producing intelligent agents that help us in our daily lives. Such agents must be able to rapidly adapt to the changing goals of their users, and the changing environments in which they operate.

These requirements lead to a balancing act that most current systems have difficulty contending with: on the one hand, human interaction and computational scalability favor the use of abstracted models of problems and environments domains; on the other hand, generating goal directed behavior in the real world typically requires accurate models that are difficult to obtain and computationally hard to reason with.

This symposium addresses the core research gaps that arise in designing autonomous systems that execute their actions in complex environments using imprecise models. The sources of imprecision may range from computational pragmatism to imperfect knowledge of the actual problem domain. Some of the research directions that this symposium aims to highlight are:

  • hierarchical approaches for goal directed autonomy in physically manifested intelligent systems (e.g., robotics)
  • formalizations for knowledge representation and reasoning under uncertainty for real-world systems and their simulations, including those based on logic as well as on probability theory
  • tradeoffs between model verisimilitude, scalability, and executability in sequential decision making
  • bridging the gaps between abstract models and reality in sequential decision making
  • online model learning and model improvement during execution
  • identifying modeling errors during plan execution
  • integrated approaches for learning representations and execution policies
  • analysis and use of abstractions in autonomous reasoning and execution

  • Symposium program available online
  • Confirmed invited speakers:
    David Aha, Naval Research Laboratory
    Emma Brunskill, Stanford University
    Jeremy Frank, NASA Ames Research Center
  • Accepted papers available online
  • Camera ready papers due Jan 29, 2018

Paper Submission

We invite paper submissions on relevant topics, which include, but are not limited to:
  • Hierarchical representation, reasoning, and planning
  • Behavior synthesis and execution in robotics
  • Planning and reasoning with abstract models while ensuring executability
  • Abstraction from controls to logic
  • Execution monitoring of autonomous systems
  • Performance evaluation of executable autonomous systems
  • Integrated task and motion planning
  • Reasoning in the presence of abstraction
  • Online model learning and model improvement
  • Detecting model errors during execution
  • Integrated representation and policy learning

We invite submissions of full papers (6-8 pages) and short/position papers (2-4 pages). We also solicit system demonstrations which highlight how some of the challenges of interest to this symposium were handled.

Papers should be submitted via the easychair portal.

Technically, symposium papers are not considered archival, and a transfer of copyright to AAAI is not required. Symposium authors are free to submit their work to other venues, but they should always check with that venue to be sure they have no problem with it.

Please register here to participate in the symposium.

Important Dates

Paper submission October 27 November 7, 2017 (extended!)
Notification November 27, 2017
Camera-ready deadline January 23, 2018

Accepted Papers [Presentation Schedule]

Download:   Single PDF  |   ZIP archive
Validation of Hierarchical Plans via Parsing of Attribute Grammars
Roman Barták, Adrien Maillard and Rafael C. Cardoso
Situated Planning for Execution Under Temporal Constraints
Michael Cashmore, Andrew Coles, Bence Cserna, Erez Karpas, Daniele Magazzeni and Wheeler Ruml
Flexible Goal-directed Agents' Behavior via DALI MASs and ASP Modules
Stefania Costantini and Giovanni De Gasperis
Creating and Using Tools in a Hybrid Cognitive Architecture
Dongkyu Choi, Son To and Pat Langley
Perspectives on the Validation and Verification of Machine Learning Systems in the Context of Highly Automated Vehicles
Werner Damm, Martin Fränzle, Sebastian Gerwinn and Paul Kröger
SiRoK: Situated Robot Knowledge - Understanding the Balance Between Situated Knowledge and Variability
Angel Daruna, Vivian Chu, Weiyu Liu, Meera Hahn, Priyanka Khante, Sonia Chernova and Andrea Thomaz
Teaching Virtual Agents to Perform Complex Spatial-Temporal Activities
Tuan Do, Nikhil Krishnaswamy and James Pustejovsky
Adversarial Regression for Stealthy Attacks in Cyber Physical Systems
Amin Ghafouri, Yevgeniy Vorobeychik and Xenofon Koutsoukos
Planning Hierarchies and their Connections to Language
Nakul Gopalan
Learning Generalized Reactive Policies using Deep Neural Networks
Edward Groshev, Aviv Tamar, Maxwell Goldstein, Siddharth Srivastava and Pieter Abbeel
Optimal LTL Planning for Multi-Valued Logics
Mohammad Hekmatnejad and Georgios Fainekos
Constraint-Based Online Transformation of Abstract Plans into Executable Robot Actions
Till Hofmann, Victor Mataré, Stefan Schiffer, Alexander Ferrein and Gerhard Lakemeyer
Learning to Act in Partially Structured Dynamic Environment
Chen Huang and Lantao Liu
Represention, Use, and Acquisition of Affordances in Cognitive Systems
Pat Langley
Learning Planning Operators from Episodic Traces
David Menager and Dongkyu Choi
Human-Agent Teaming as a Common Problem for Goal Reasoning
Matthew Molineaux, Michael Floyd, Dustin Dannenhauer and David Aha
Interaction and Learning in a Humanoid Robot Magic Performance
Kyle Morris, John Anderson, Meng Cheng Lau and Jacky Baltes
A Framework for Complementary Video-Game Companion Character Behavior
Gavin Scott and Foaad Khosmood
Reasoning About Domains with PDDL
Alexander Shleyfman and Erez Karpas
On Chatbots Exhibiting Goal-Directed Autonomy in Dynamic Environments
Biplav Srivastava
Safe Goal-Directed Autonomy and the Need for Sound Abstractions
Siddharth Srivastava
Exploiting Micro-Clusters to Close The Loop in Data-Mining Robots for Human Monitoring
Einoshin Suzuki
Robot Behavioral Exploration and Multi-modal Perception using Dynamically Constructed Controllers
Suhua Wei, Shiqi Zhang, Jivko Sinapov, Jesse Thomason and Peter Stone
Learning Abstractions by Transferring Abstract Policies to Grounded State Spaces
Lawson L.S. Wong
Information-Efficient Model Identification for Tensegrity Robot Locomotion
Shaojun Zhu, David Surovik, Kostas Bekris and Abdeslam Boularias

Invited Talks

An Integrated Model of Goal Reasoning
David Aha, U.S. Naval Research Laboratory

Goal reasoning (GR) allows an agent to dynamically reason about the relative utilities of goals it could pursue, which may result in changing its active goals. We have studied processes that support GR, focusing on situation assessment and decision making, and its application in (simulated and real) deliberative control tasks. I will present a simple integrated model for GR, review a progression of our work and how it relates to SIRLE’s themes, and summarize some of our current research objectives.

Dr. David Aha leads the Adaptive Systems Section within NRL’s Navy Center for Applied Research in AI. His research interests include intelligent (e.g., goal reasoning) agents, planning, case-based reasoning, explainable AI (XAI), machine learning, and related topics. He co-organized 35 international events related to these topics (e.g., ICCBR-17, AAAI-18 Senior Member Track, FAIM-18 XAI Workshop), launched the UCI Repository for ML Databases, served as a AAAI Councilor, co-created the AAAI Video Competition, received 5 publication awards, and gave the IAAI-17 Engelmore Memorial Lecture. He has led 4 DARPA (e.g., XAI) or ONR evaluation teams. His group regularly hosts post-doctoral researchers and many summer visitors.

From Prior Experience to Future Plans Using Models and Hierarchy
Emma Brunskill, Stanford University

The Challenge of Verification and Validation of Automated Planning Systems [Slides]
Jeremy Frank, NASA Ames Research Center

We discuss the opportunities for autonomous systems to perform reflection on their planners by adapting the models used to build plans. We first describe model-based planning systems, a form of automated planning system driven by declarative models of the planning domain. These models include descriptions of the conditions and effects of actions on the state of the world. When planning the activities of cyber-physical systems, the command and data representation of the system must be formally abstracted to the actions and states described in the planning system model. When the execution of a plan either fails or produces unexpected outcomes, the execution trace can be abstracted and compared to the predicted state according to the planning model, producing a list of discrepancies; these discrepancies can then be used to fix the model. This provides part of a reflection capability, namely, a set of well-formed problems with the domain model, the abstractions, or both. The challenge lies in the rest of the reflection capability, namely, a set of techniques for changing the models or the abstractions. We discuss these challenges and describe some of the options for addressing them.

Dr. Jeremy Frank is the many-times great grandson of the infamous Dr. Victor Frankenstein; you could say that Artificial Intelligence runs in the family.  While unable to completely live down the infamous family name, Dr. Frank has nevertheless managed to bury the most unsavory parts of his history, and has made a minor name for himself by writing some obscure papers in the areas of computer science and AI, while moonlighting as a rocket scientist for a little-known space exploration agency that resides within a bloated, Byzentine bureacracy that poorly serves a large Western hemisphere country. Dr. Frank strives one day to abandon these fruitless pursuits and give over his time to his passions of writing fiction, birding, and cooking, but in the meantime, has shown some small skill in organizing people and fundraising.

In other words, Dr. Jeremy Frank is the Group Lead of Planning and Scheduling Group, in the Intelligent Systems Division, at NASA Ames Research Center. He received his Ph. D. from the Department of Computer Science, at the University of California at Davis, in June 1997. He also has a B.A. in Mathematics from Pomona College. Dr. Frank’s work involves the development of automated planning and scheduling systems for use in space mission operations, the integration of technologies for planning, plan execution, and fault detection for space applications, and the development of technology to enable astronauts to autonomously operate spacecraft. Dr. Frank has published over 50 conference papers, nine journal papers, and three book chapters, and received over 40 NASA awards, including the Exceptional Achievement Medal, the Silver Snoopy, and the NASA Engineering and Safety Center Award.

(Thanks to Jeremy for offering an alternative bio!)

Organizing Committee

Siddharth Srivastava Arizona State University
Shiqi Zhang Cleveland State University
Nick Hawes University of Oxford
Erez Karpas Technion – Israel Institute of Technology
George Konidaris Brown University
Matteo Leonetti University of Leeds
Mohan Sridharan The University of Auckland
Jeremy Wyatt University of Birmingham

Program Committee

Christopher Amato Northeastern University
J. Benton NASA Ames Research Center / AAMU-RISE
Joydeep Biswas University of Massachusetts Amherst
Minh Do NASA Ames Research Center
Esra Erdem Sabanci University
Georgios Fainekos Arizona State University
Alberto Finzi Universita' di Napoli Federico II
Michael Gelfond Texas Tech University
Marc Hanheide University of Lincoln
Laura Hiatt U.S. Naval Research Laboratory
Luca Iocchi Sapienza University of Rome
Leslie Kaelbling MIT
Sven Koenig University of Southern California
Lars Kunze University of Oxford
Bruno Lacerda University of Oxford
Gerhard Lakemeyer RWTH Aachen University
Daniele Magazzeni King's College London
Lenka Mudrova University of Birmingham
Tim Niemueller RWTH Aachen University
Andrea Orlandini National Research Council of Italy (ISTC-CNR)
Federico Pecora Orebro University
Subramanian Ramamoorthy The University of Edinburgh
Mark Roberts Naval Research Laboratory
Alessandro Saffiotti Orebro University
Enrico Scala ANU Research School in Computer Science
Jivko Sinapov Tufts University
Sylvie Thiebaux ANU
Yu Zhang Arizona State University
Shlomo Zilberstein University of Massachusetts Amherst