[Event at CIG] 1st CLaRAMAS workshop at AAMAS 2026

[Stefano Mariani]

Dear Colleague,
we hope this email finds you well :)

We cordially invite you to submit your research work to the /1st 
International Workshop on “Causal Learning and Reasoning in Agents and 
Multiagent Systems”/ (CLaRAMAS), hosted by the 25th International 
Conference on Agents and Multiagent Systems (AAMAS).

The workshop seeks contributions at the *crossroads between learning of 
and reasoning with causal models, and engineering agents and agent-based 
systems* (both programming and learning approaches).
Check out the workshop’s full scope & aims and the Call for Papers (also 
reported at the end of this email) on the workshop website: 
https://claramas-workshop.github.io/claramas2026/

/The submission deadline is February, 4th 2026./

Looking forward to receive your contributions,
our best regards.

The CLaRAMAS Program Chairs
    --- /Stefano Mariani, Mehdi Dastani, André Meyer-Vitali, Julien Siebert/

##### Call for Papers #####

The concept of an “agent” represents a foundational abstraction in 
software engineering, encapsulating the notion of *agency*—namely, the 
capacity of a software entity to bring about effects in pursuit of 
specific goals within its operating environment. Exercising agency 
requires the ability to interpret the structure and dynamics of that 
environment and to anticipate its responses to the agent’s actions. In 
essence,

/agency hinges on understanding and leveraging the causal relationships 
among observable and controllable variables (e.g., through sensors and 
actuators)./

Such causal reasoning is indispensable for planning actions that 
reliably achieve intended objectives—a principle reinforced by recent 
research on causal inference in emerging “agentic AI” systems.

This requirement extends naturally to multi-agent systems (MAS), a 
cornerstone of distributed artificial intelligence, where multiple 
agents coexist and interact within a shared environment. These 
interactions – whether cooperative or competitive – contribute to 
individual and systemic goals, either through direct communication or 
indirect influence on the environment. Consequently,

/effective coordination in multi-agent settings depends on a causal 
understanding of interdependencies among agents’ behaviors./

Only by modeling these reciprocal influences can agents achieve robust 
and purposeful collaboration (or competition) toward their respective 
objectives.

⚠️ /However, this fundamental role of causal modelling of the 
agent-environment and agent-agent relationships is not yet widely and 
deeply discussed in the AAMAS community./ ⚠️

Accordingly, CLaRAMAS welcomes submissions dealing with the following 
*topics of interest*:

  * how to integrate causal learning in agent architectures and MAS
  * how to carry out causal learning of agent-environment and
    agent-agent relationships from the standpoint of an individual agent
    or of the MAS as a whole
  * how causal modelling and learning can be integrated in
    agent-oriented software engineering methodologies
  * exploring causal explainability, safety, and robustness in agent(s)
    design, for instance in robotic and multi-robot systems
  * how causal learning may integrate with learning-based approaches to
    agent design, such as with Reinforcement Learning for counterfactual
    reasoning, credit assignment, policy evaluation, policy improvement
    in single- and multi-agent systems
  * theoretical foundations of causal learning and reasoning in single-
    and multi-agent systems, including the relationship between
    sequential decision making (e.g., MDPs) and Pearl structural causal
    models, integration of game-theoretic formalisms, etc.
  * practical applications of causal learning and reasoning in MAS
  * cooperative planning, prediction, and diagnosis using (perhaps,
    partially) shared causal models
  * combining causal learning and reasoning with planning and adaptive
    control, including model-based, model-free and hybrid approaches
  * cooperative causal discovery and inference in MAS
  * neuro-symbolic AI via causal models
  * evaluation and benchmarks for causal MAS applications, including
    datasets, metrics, simulators, and reproducible experimental pipelines

Check out the /submission dates and instructions/ at the CLaRAMAS 
website: https://claramas-workshop.github.io/claramas2026

##### #####


-- 
------------------------------
Stefano Mariani, PhD
Tenure-track researcher (RTDb) @ University of Modena and Reggio Emilia
   >https://smarianimore.github.io
------------------------------
<<< Well being notice >>>
I care about your work-life balance: I sent this email at a time that works for me; please, read and reply at a time that works for you!
:)
------------------------------