Conchware
Hybrid-Casual puzzle developer
&
Heterogeneous Autonomous Divergent Swarm Intelligence (H.A.D.S.I.) R&D
What is a Heterogeneous Autonomous Divergent Swarm Intelligence (H.A.D.S.I.)?
A Heterogeneous Autonomous Divergent Swarm Intelligence (H.A.D.S.I.) is a "Limited Memory AI" that enables a group of identically programmed heterogeneous drones to autonomously seek solutions to complex problems in real-time, while at the same time diverging as individuals and groups to carry out separate tasks from the collective swarm based on localized conditions or swarm combinations.
This is achieved without reliance on weighted neural node networks, effectively circumventing challenges associated with 'Generative A.I. Hallucinations', 'Reversal Curses', and 'LLM Sleeper Agents', and the significant costs of training Large Language Models that entails.
Below are some examples of the projects we've developed during our extensive testing and research:
Non-weighted neural node Autonomous Swarm Intelligence (A.S.I.)
S.I. for Foraging, Gathering, Scouting, and Defending Purposes
Introducing a real-time strategy (R.T.S.) simulation featuring two factions, Red vs. Blue, engaging in foraging, gathering, scouting, and base defense autonomously, without direct player intervention.
Drawing inspiration from the foundational concepts of Total Annihilation and Defense of the Ancients (DotA), we've crafted this prototype with the future of RPG and MOBA gaming in mind.
S.I. for Resource Allocation, Logistics and Transportation, and Factory and Warehouse Operations
This simulation offers a comprehensive view of the operational journey, spanning from the production of base materials to the delivery of finished products to consumers.
Within the simulation, items (represented by letters) are maneuvered by two human agents (depicted as 🙂 ), guided by Conchware’s S.I. on resource allocation (which resources should go to which station first) for optimal efficiency. This guidance entails determining the priority sequence for allocating resources to stations, ensuring a streamlined process from production to end consumers.
The demo comprises four layers of stations:
* Base materials (A).
* First-refined resources (B, C, and D): each requiring 3 units of A to produce one of them.
* Second-level goods or Finished Products (E, F, G, H, I, and J): each needing 1 unit of their lower-tier resources (both Base material and First-refined resources) to make the final product.
For example, E necessitates 1 unit of A, B, C, and D, while J requires 1 unit of A, B, C, D, E, F, G, and I.
* Consumers and/or their households ( 🙂 )