Skip to content

[TASK] Explore conversational consensus algorithms #5

Description

@MrCsabaToth

Agents can malfunction and mechanisms need to be explored how the collective system can endure outages, elect spare agents in place of malfunctioned agents, rule out adversaries and survive attacks.

Describe the solution you'd like
This is deep research and would possibly need game theory and mathematical logic approaches with the extra challenge of LLMs being stochastic.

Describe alternatives you've considered
Combining specialized tools to assist with these challenges (especially once Koog will provide full ReAct and agentic capabilities).

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

enhancementNew feature or request

Type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions