AI lib, PsychSim/Thespian

[DrAltaica]

I'm trying to make AI library/WorldForge client based on Thespian

Decision-theoretic models have become increasingly popular as a basis for solving agent and multiagent problems, due to their ability to quantify the complex uncertainty and preferences that pervade most nontrivial domains. However, this quantitative nature also complicates the problem of constructing models that accurately represent an existing agent or multiagent system, leading to the common question, “Where do the numbers come from?” In this work presents a method for exploiting knowledge about the qualitative structure of a problem domain to automatically derive the correct quantitative values that would generate an observed pattern of
agent behavior. In particular, we propose the use of piecewise linear functions to represent probability distributions and utility functions with a structure that we can then exploit to more efficiently compute value functions. More importantly, we have designed algorithms that can (for example) take a sequence of actions and automatically
generate a reward function that would generate that behavior within our agent model. This algorithm allows us to efficiently fit an agent or multiagent model to observed behavior.



Our decisions to act are influenced by how we believe others will react. Whether we believe a message depends not only on its content but also on our model of the communicator. Giving its importance in human social interaction, modeling theory of mind can play a key role in enriching social simulations.

PsychSim, a social simulation tool, operationalizes existing psychological theories as boundedly rational computations to generate more plausibly human behavior. PsychSim allows a user to quickly construct a social scenario where a diverse set of entities, groups or individuals, interact and communicate. Each entity has its own preferences, relationships (e.g., friendship, hostility, authority) with other entities, private beliefs, and mental models about other entities. The simulation tool generates the behavior for these entities and provides explanations of the result in terms of each entity’s preferences and beliefs. The richness of the entity models allows one to explore the potential consequences of minor variations on the scenario.


A central aspect of the PsychSim design is that agents have fully specified decision-theoretic models of others. Such quantitative recursive models give PsychSim a powerful mechanism to model a range of factors in a principled way. For instance, we exploit this recursive modeling to allow agents to form complex attributions about others, enrich
the messages between agents to include the beliefs and preferences of other agents, model the impact such recursive models have on an agent’s own behavior, model the influence observations of another’s behavior have on the agent’s model of that other, and enrich the explanations provided to the user. The decision-theoretic models in particular give our agents the ability to judge degree of credibility of messages in a subjective fashion that can consider the range of influences that sway such judgments in humans


Pynsdath has released some code under the MIT licence on github https://github.com/pynadath/psychsim but it is missing documintation and I'm not sure it' even complete

It's not a vary big project being just under 1 MB of Python3.

I tried Email Pynadath but didn't get a reply and with my luck he's gotten some terminal disease or died. So right now I'm in need of someone familiar with Python to reverse engineer the code and help me write docs and examples of how to use it.


youtu.be/3tlcSw2LmLY

[fluffrabbit]

tl;dr

From what little I absorbed of the OP, I think the focus on multi-agent systems is odd. What about reinforcement learning? What do you do for situations where the player is alone in a room with a single entity? They've got to perform convincingly as individuals, and a lot of behaviors would have to be hard-coded if you're going for something like Morrowind. Not sure I follow the theory.

[DrAltaica]

tl;dr

From what little I absorbed of the OP, I think the focus on multi-agent systems is odd.

Multi Agent system is just what Object Oriented in known as outside programming.

What about reinforcement learning?

As far as I know RL is still an open problem.

What do you do for situations where the player is alone in a room with a single entity?

No. Multi agent does mean that you simulate a bunch of agent as one collective mass, like Lanchester’s Laws, it means you solve your problem by making a lot of independent agents that when they interact solve your problem.like this They've got to perform convincingly as individuals, and a lot of behaviors would have to be hard-coded if you're going for something like Morrowind. Not sure I follow the theory.[/quote] once you can predict what other chars want a lot of stuff becomes easy.

[DrAltaica]

PyschSim is basically the oposite of reinforcement learning. you give it a path and it gives you the reward function.

[fluffrabbit]

PyschSim is basically the oposite of reinforcement learning. you give it a path and it gives you the reward function.

That makes sense. Like the Office Assistant. "It looks like you're trying to write a suicide note. Can I help you with that?"

There is no README in that repo. Either find a paper for it or move on; it won't explain itself.