Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Poker Tournament for LLMs

pokerbattle.ai

Post reply on HN
Print view View original

151–160 of 212 posts

Re: Poker Tournament for LLMs

#151

Post by catketch »

sceptic123 wrote: post #134
tarruda wrote: post #98

Earlier quoted context omitted.

> LLMs do not have a mechanism for sampling from given probability distributions Would a LLM with tool calls be able to do this?

Then it's not the LLM doing the work

this is is a distinction without a difference in many instances. I can easily ask an llm to write a python tool to produce random numbers for a given distribution and then use that tool as needed. The LLM writes the code, and uses the executable result. Then end black box result is the LLM doing the work

Re: Poker Tournament for LLMs

#152

Post by sceptic123 »

catketch wrote: post #151
sceptic123 wrote: post #134

Earlier quoted context omitted.

Then it's not the LLM doing the work

this is is a distinction without a difference in many instances. I can easily ask an llm to write a python tool to produce random numbers for a given distribution and then use that tool as needed. The LLM writes the code, and uses the executable result. Then end black box result is the LLM doing the work

But why limit it to generating random numbers, isn't the logical conclusion that the LLM writes a poker bot instead of playing the game? How would that demonstrate the poker skills of an LLM?

Re: Poker Tournament for LLMs

#153

Post by Lerc »

48terry wrote: post #146
Lerc wrote: post #87

Earlier quoted context omitted.

>3) LLMs do not have a mechanism for sampling from given probability distributions. E.g. if you ask LLM to sample a random number from 1 to 10, it will likely give you 3 or 7, as those are overrepresented in the training data. I am not sure that is true. Yes it will likely give a 3 or 7 but that is because it is trying to represent that distribution from the training data. It's not trying for a random digit there, it…

I have a better idea: random.randint(1,10)

That requires tool use or some similar specific action at inference time.

The technique I suggested would, I think, work on existing model inference methods. The ability already exists in the architecture. It's just a training adjustment to produce the parameters required to do so.

Re: Poker Tournament for LLMs

#155

Post by recursive »

gsinclair wrote: post #36
michalsustr wrote: post #31

I have PhD in algorithmic game theory and worked on poker. 1) There are currently no algorithms that can compute deterministic equilibrium strategies [0]. Therefore, mixed (randomized) strategies must be used for professional-level play or stronger. 2) In practice, strong play has been achieved with: i) online search and ii) a mechanism to ensure strategy consistency. Without ii) an adaptive opponent can learn to exp…

FWIW, I’d bet some coin that current CharGPT would provide a genuine pseudo-random number on request. It now has the ability to recognise when answering the prompt requires a standard algorithm instead of ordinary sentence generation. I found this out recently when I asked it to generate some anagrams for me. Then I asked how it did it.

I don't think LLMs can reliably explain how they do things.

Re: Poker Tournament for LLMs

#156

Post by ramoz »

michalsustr wrote: post #31

I have PhD in algorithmic game theory and worked on poker. 1) There are currently no algorithms that can compute deterministic equilibrium strategies [0]. Therefore, mixed (randomized) strategies must be used for professional-level play or stronger. 2) In practice, strong play has been achieved with: i) online search and ii) a mechanism to ensure strategy consistency. Without ii) an adaptive opponent can learn to exp…

An LLM in a proper harness (agent) can do all of those things and more.

Re: Poker Tournament for LLMs

#157

Post by akd »

michalsustr wrote: post #31

I have PhD in algorithmic game theory and worked on poker. 1) There are currently no algorithms that can compute deterministic equilibrium strategies [0]. Therefore, mixed (randomized) strategies must be used for professional-level play or stronger. 2) In practice, strong play has been achieved with: i) online search and ii) a mechanism to ensure strategy consistency. Without ii) an adaptive opponent can learn to exp…

Facebook built a poker bot called Pluribus that consistently beat professional poker players including some of the most famous ones. What techniques did they use?

https://en.wikipedia.org/wiki/Pluribus_(poker_bot)

Re: Poker Tournament for LLMs

#158

Post by furyofantares »

michalsustr wrote: post #31

I have PhD in algorithmic game theory and worked on poker. 1) There are currently no algorithms that can compute deterministic equilibrium strategies [0]. Therefore, mixed (randomized) strategies must be used for professional-level play or stronger. 2) In practice, strong play has been achieved with: i) online search and ii) a mechanism to ensure strategy consistency. Without ii) an adaptive opponent can learn to exp…

> 3) LLMs do not have a mechanism for sampling from given probability distributions. E.g. if you ask LLM to sample a random number from 1 to 10, it will likely give you 3 or 7, as those are overrepresented in the training data.

You can have them output a probability distribution and then have normal code pick the action. There's other ways to do this, you don't need to make the LLM pick a random number.

Re: Poker Tournament for LLMs

#159

Post by 48terry »

zie1ony wrote: post #111

Hi there, I'm also working on LLMs in Texas Hold'em :) First of all, congrats on your work. Picking a form of presenting LLMs, that playes poker is a hard task, and I like your approach in presenting the Action Log. I can share some interesting insights from my experiments: - Findin strategies is more interesting than comparing different models. Strategies can get pretty long and specific. For example, if part of the…

Question: What makes LLMs well-suited for the task of poker compared to other approaches?

Re: Poker Tournament for LLMs

#160

Post by jgalt212 »

akd wrote: post #157
michalsustr wrote: post #31

I have PhD in algorithmic game theory and worked on poker. 1) There are currently no algorithms that can compute deterministic equilibrium strategies [0]. Therefore, mixed (randomized) strategies must be used for professional-level play or stronger. 2) In practice, strong play has been achieved with: i) online search and ii) a mechanism to ensure strategy consistency. Without ii) an adaptive opponent can learn to exp…

Facebook built a poker bot called Pluribus that consistently beat professional poker players including some of the most famous ones. What techniques did they use? https://en.wikipedia.org/wiki/Pluribus_(poker_bot)

> Pluribus, the AI designed by Facebook AI and Carnegie Mellon University to play six-player No-Limit Texas Hold'em poker, utilizes a variant of Monte Carlo Tree Search (MCTS) as a core component of its decision-making process.
Post reply on HN