HSML + Active Inference

Join VERSES CEO, Gabriel René; Chief Scientist, Dr. Karl Friston and Director of Research, Maxwell Ramstead as they shed light on the power of HSML and its significance as a universal data language, enabling seamless interoperability and intelligent reasoning among diverse agents.
Connect with us 🕸️:
➤ Twitter: / helloverses
➤ LinkedIn: / helloverses
➤ Instagram: / helloverses
➤ Facebook: / helloverses
Join us in building a Smarter World:
www.verses.ai/
#interoperability #intelligence #AI #artificalintelligence #HSML #XAI #activeinference

[@anIntelligentAgent]

Very exciting to see Dr. Friston's work translated to this practical application. To pursue the analogy somewhat: I assume that an IA (intelligent agent) in this system is storing its beliefs about the world and other agents in the form of (nested?) HSML (as brains store beliefs using action potentials and structured neuronal graphs), and that it can also share its beliefs with other IA in the HSML format (as neurons communicate with other neurons or as organisms communicate with each other using shared language). The IA could be a single neuron or a complete brain in this analogy, and HSML either action potentials or shared language? Facinating stuff.
The active inference part confuses me somewhat. What drives the agents to learn more about the world? Who or what encodes their beliefs about what needs to be done or optimised? Could IA's share beliefs that go beyond their intended use? (philosophizing with other IA's about the meaning of their exisitence instead of optimising warehouse routes) Could networks of IA's be considered entirely new IA's in the same way networks of neurons constitute the nervous systems of discrete organisms?

[@princee9385]

Perfectly put 🎉