#Вакансия #машиное_обучение #python #удаленно
Занятость: удаленная работа. фриланс
Бюджет: 8.000 - 10.000$
Необходимые навыки:
- Python, C++, ML
- Знание игры в покер
Деадлайн: 2 месяца
Implementation of the key ideas from the Pluribus poker AI that plays Texas Hold'em PokerAI iteration.
Iterate on the AI algorithms and the integration into the poker engine.
Integrate the AI strategy to support self-play in the multiplayer poker game engine. In the game-engine, allow the replay of any round the current hand to support MCCFR.
Implement the creation of the blueprint strategy using Monte Carlo CFR minimization. Add the real-time search for better strategies during the game
It works with image recognition, montecarlo simulation and a basic MCCFR. algorithm.
the bot will be outside the VM and be able to read client(inside the VM)..the bot will be able to read vm hands
to visualise the AI as it plays, and debug particular situations visually. The idea is to use a live web-visualisation server like TensorBoard, so i will just push the poker game state, and it will be reflected in the visualisations, so you can see what the agents are doing.
it just has to response to ocr the action which i am gonna play manualy
ocr goona read the current poker table situation. make a request to ai what to do. ai would make calculations < 13 sec and response with the best action
call, raise X bb, fold
Whitepapers:
https://science.sciencemag.org/content/sci/suppl/2019/07/10/science.aay2400.DC1/aay2400-Brown-SM.pdfhttp://poker.cs.ualberta.ca/publications/NIPS12.pdfhttps://icml.cc/media/Slides/icml/2019/102(11-11-00)-11-12-15-4443-deep_counterfac.pdfTo assess the compression of the dispersion of agents' actions, Pluribus uses AIVAT (A new Variance Reuction Technique). It is necessary to develop pseudocode and write an algorithm based on it.
WP:
https://arxiv.org/pdf/1612.06915.pdf@clowninghost