IJCAI-17 Workshop on
Architectures for Generality & Autonomy

Architectures for Generality & Autonomy

Call for Papers

How can we make machines that are more self-sufficient and autonomous than possible today? How can we make machines that can achieve complex tasks, even when given only vague, high-level instructions? How can specialized AI technologies fit together to form architectures with increased autonomy and general intelligence?

Recent developments in the field, IJCAI-17's Special Theme on Autonomy, and its co-location with the Tenth Conference on Artificial General Intelligence, have prompted us to organize the First International Workshop on Architectures for Generality & Autonomy (AGA 2017) at the 26th International Joint Conference on Artificial Intelligence. AGA 2017 welcomes a wide range of people who are interested in the "grand AI dream" of creating general or "strong" AI. We aim to bring together researchers from different AI sub-disciplines and communities in order to encourage cross-disciplinary collaboration, including those from the IJCAI, AGI ("artificial general intelligence"), and BICA ("biologically inspired cognitive architectures") Societies, as well as cognitive science, neuroscience, philosophy, and computer science.

While many recent advances (e.g. deep learning techniques and hierarchical reinforcement learning) fulfill the requirement of continued advances and success in AI, their ability to bring forth human-level artificial intelligence remains unclear. The workshop will focus on AI and cognitive architectures with an eye towards artificial general intelligence, autonomy and task-independent computational theories about how (natural and artificial) minds acquires knowledge and skills.

Some scientific and methodological questions and topics this workshop aims to address (but is not limited to) include:

  • Is general intelligence necessary to achieve full autonomy in an artificial system?
  • How can “narrow AI” techniques be used towards artificial general intelligence?
  • What differentiates generally intelligent systems from specialists and “idiot savants”?
  • Do autonomy and general intelligence depend on the same underlying principles — is autonomy supervenient on general intelligence — or are these orthogonal dimensions?
  • Is the separation between general and specialized intelligence sharp or gradual?
  • Do we need radically different approaches than those in use today, to build systems with human-level intelligence?
  • How can current state of the art methods in AI address human-level AI?
  • Do we need special programming languages to implement reflective (self-referential) / creative / generally intelligent systems?
  • Does general intelligence depend on underlying core principles, or does generality simply result from the combination of a large number of specialized functions?
  • Can general AI be achieved through hand-crafted architectures or must it emerge through self-organizing (constructivist) principles?

Key Information:

August 19
RMIT Building 80, floor 9, room 80.09.10, 445 Swanston Street, Melbourne, Australia

Paper & Demo submission:

Due date:
May 23
Notification date:
June 17
Camera-ready date:
July 28
Submission system:


AGI-17 Conference
August 15-18
EGPAI 2017 Workshop
August 20




Time Event
8:55-9:20 Dr. Kristinn R. Thórisson, Dr. Pei Wang and Dr. Joscha Bach
Architectures for Generality & Autonomy — A View From Past to Present
9:20-10:00Invited talk: Dr. Aaron Sloman (video presence)
Why can't (current) machines reason like Euclid or even human toddlers?
10:00-10:30Coffee break
10:30-12:30Contributed talks
  • Joscha Bach
    The Cortical Conductor Theory: Towards Addressing Consciousness in AI Models (paper)
  • Hiroshi Yamakawa, Naoya Arakawa and Koichi Takahashi
    Reinterpreting The Cortical Circuit (paper | slides)
  • Claes Strannegård, Nils Svangård, David Lindström, Joscha Bach and Bas Steunebrink
    The Animat Path to Artificial General Intelligence (paper | slides)
  • Bernhard Hengst, Maurice Pagnucco, David Rajaratnam, Claude Sammut and Michael Thielscher
    Context In Cognitive Hierarchies
    (paper | slides | supporting videos available upon request)
  • Noa Agmon, Oleg Maximov, Ariel Rosenfeld, Shai Shlomai and Sarit Kraus
    Multiple Robots For Multiple Missions: Architecture for Complex Collaboration
    (paper | slides | supporting video [YouTube])
14:00-16:00 Contributed talks
  • Louise Dennis, Elisa Cucco and Michael Fisher
    A General Architecture for Flexible Autonomous Systems
    (paper | slides | supporting video available upon request)
  • Pawel Gomoluch, Dalal Alrajeh, Alessandra Russo and Antonio Bucchiarone
    Towards learning domain-independent planning heuristics (paper | slides)
  • Gavin Rens, Thomas Meyer and Deshendran Moodley
    A Stochastic Belief Management Architecture for Agent Control (paper | slides)
  • Patrick Hammer, Tony Lofthouse and Pei Wang
    OpenNARS Demonstration of Autonomous Learning and Decision-Making (paper | slides)
  • Kristinn R. Thórisson
    Machines with Autonomy & General Intelligence: Which Methodology? (paper)
16:00-16:30Coffee break
16:30-17:30 Panel Discussion / Brainstorming Session
17:30-18:00 Poster & Demo Session

Invited speaker:

Dr. Aaron Sloman

University of Birmingham


Papers were accepted after being peer reviewed by 2-3 reviewers per paper.


Dr. Kristinn R. Thórisson (contact)

Reykjavik University & Icelandic Institute for Intelligent Machines, Iceland

Dr. Pei Wang

Temple University, U.S.

Dr. Kamilla R. Jóhannsdóttir

Reyjkavik University, Iceland

Dr. Joscha Bach

Harvard, U.S.

Jordi Bieger, M.Sc.

Reykjavik University, Iceland

Program Committee

Joscha Bach Harvard University, U.S.
Tarek Richard Besold University of Bremen, Germany
Jordi Bieger Reykjavik University, Iceland
Antonio Chella University of Palermo, Italy
Haris Dindo Yewno & University of Palermo, Italy
Patrick Hammer Graz University, U.S.
Helgi P. Helgason Activity Steeam, Iceland
Kamilla R. Jóhannsdóttir Reykjavik University, Iceland
Jan Koutnik NNAISENSE, Switzerland
David Kremelberg Icelandic Institute for Intelligent Machines, Iceland
Kai-Uwe Kühnberger University of Osnabrück, Germany
Richard Mallah Future of Life Institute, U.S.
Ramon Lopez de Mantaras AI Research Institute or the Spanish National Research Council, Spain
Laurent Orseau DeepMind, U.K.
Paul Rosenbloom University of Southern California, U.S.
Ricardo Sanz Universidad Politécnica de Madrid, Spain
Jürgen Schmidhuber IDSIA, Switzerland
Bas Steunebrink NNAISENSE, Switzerland
Claes Strannegård University of Gothenburg, Sweden
Kristinn R. Thórisson Reykjavik University & Icelandic Institute for Intelligent Machines, Iceland
Pei Wang Temple University, U.S.


The central focus of AGA 2017 is autonomous and intelligent general-purpose systems. We are interested in hearing a diverse set of views from researchers in numerous fields —

  • Artificial intelligence
  • Cognitive science & psychology
  • Neuroscience
  • Philosophy
  • Computer science
  • Engineering

— with an aim to discuss a wide range of methods, assumptions, approaches, and systems under development.

We welcome —
  • technical papers
  • overview / summary papers
  • application papers
  • position papers
  • demonstrations

— on a wide range of topics related to achieving the grand dream of AI research, including but not limited to:

  • Analysis of requirements for autonomy / generality
  • Design proposals for cognitive architectures targeting generality and/or autonomy
  • Synergies between AI approaches with an eye towards AGI
  • New programming languages relevant to generality and autonomy
  • New methodologies relevant to generality and autonomy
  • New architectural principles relevant to generality and autonomy
  • New theoretical insights relevant to generality and autonomy
  • New ways to use current methodologies to achieve autonomy / generality
  • Analysis of the potential and limitations of existing approaches
  • Methods for epigenetic development
  • Education to achieve general or (super)human performance
  • … and much more

Papers should be between 2 and 12 pages (excluding references) and describe the authors' original work in full (no extended abstracts). Formatting Guidelines, LaTeX Styles and Word Template can be dowloaded from here. Papers will be subjected to peer-review and can be accepted for oral presentation and/or poster presentation. For papers that have previously been submitted to IJCAI and rejected, we ask authors to append the reviews and their responses to aid our review process.

Proposals for Demonstrations should be accompanied with a 2-page description for inclusion in the workshop's pre-proceedings. Examples include, but are not limited to: (interactively) demonstrating the performance of a robot, (cognitive) architecture, or design methodology.

Oral presentations should be given by one of the authors during one of the Contributed Talks Sessions. Posters and demonstrations will be presented during the Demonstrations and Posters Session.

The paper submission deadline has passed. Authors can access submitted material on EasyChair using this link

Accepted papers will be gathered into a volume of pre-proceedings and published on this website before the workshop. We are looking into the possibility of producing a special issue for an archival journal.