My name is Arseniy Pertzovsky and this page presents my GitHub projects in a structured way. The main fields of my work are Multi-agent Systems (MAS), Search AI, RL, ROS, MARL, ML, WEB, IOT. The projects are divided into sections for the sake of convenience.
- DCOP | CAMS, Max-Sum_MST, DSA_MST, CADSA, DSSAπ
- MARL | MADDPG π (PPO in MA setting)
- MARL | FedRL
- MARL | ae_comm
- MAPF | CA*, CBS, DSA_MAPF, MGM_MAPF
TODO:
MAPPO, QMix, ROMA, COMA, Value Decomposition, MF-Q, MF-AC, MAAC, DBS-DQN, DGN, MASAC, MATD3, QTRAN, MULTI-AGENT AUTOCURRICULA, IQL, TarMAC, SEAC, BiCNet
- Learning Multi-Armed Bandits
- Learning Dynamic Programming (Policy Iteration, Value Iteration)
- Learning Monte-Carlo RL
- Learning TD-Learning
- DQN (variant 2, variant 3)
- REINFORCE (variant 2, variant 3)
- A2C (A3C)
- PPO (variant 2)
- DDPG (variant 2)
- SAC
TODO:
I2A, TD3, PPG, HER, POLO, MuZero
TODO:
Genetic Algorithms, GANs, Regression, Logistic Regression, K-Nearest Neighbors, Naive Bayes, Support Vector Machines (SVM), Monte Carlo Tree Search (MCTS) (source 1), Decision Tree, Random Forest, AdaBoost, Gradient Boost, CatBoost, XGBoost, LightGBM, Graph NN (source 1 - DGL, source 2)
- RRT Implementation - Generic Version
- A* Implementation - Generic Version
- A* Simulator - Pathfinding
- Learning Topological Sorting
- MAPF-LNS2 (+SIPPS), PrP (+SIPPS), SIPPS Implementations in Python
- LaCAM (+PIBT), LaCAM* (+PIBT), PIBT Implementations in Python
- MAS simulator
- DCOP | Toy DCOP Max-sum Simulation
- DCOP | Simulator DCOP_MST (version 5) (prev: (version 1), (version 2). (version 3), (version 4))
- DCOP | Project of Ben Rahmut
- DCOP | DCOP-MDP Simulator
- DCOP | Async-DCOP_MST Simulator version 2 (with Ben) (prev: (version 1 - incorrect))
- MAPF | MAPF simulator (version 2) (prev: (version 1))
- Gentleman_Algorithm (MAPF)
- Vanila APFs for MAPF | Potential Fields in MAPF
- APFs for MAPF & LMAPF | Artificial Potential Fields in MAPF and Lifelong MAPF (version 2) (prev: version 1)
- CGA, SACG, CGA-LMAPF | Corridor-Generating Algorithm and Single-Agent Corridor-Generating problem
- CGA-MAPF | Corridor-Generating Algorithm for MAPF (version 2)(prev: version 1)
- ROS | ROS Package for Hamster Robots
- Search | Voronoi + A* + RRT (robot navigation)
- DCOP | Implementation of Max-Sum_MST in ROS Platform
- DCOP | Implementation of CAMS in ROS Platform
- Current: ML For Trading - Drafts
- Stocks Simulator (Streamlit + Matplotlib)
- Learning to implement NN on Stocks
- GA in stocks
- Learning from "Advances in Financial ML" book
- Learning from "ML for Trading" book
- (Trading_model_first_trying), (streamlit app 1), (streamlit app 2), (website - flask), (website - react), (Stocks Gym Env)
- Learning Git and GitHub
- Learning PyTorch
- Learning ROS Essentials
- Learning PL
- Learning SimPy
- Learning Tkinter
- Learning Matplotlib
- Learning Plotly
- Learning Pandas
- Learning Neptune.ai
- Learning Gradio.app
- Learning Pygame 1
- Learning Pygame 2
- Learning SQLite
- Learning Gym (OpenAI)
- Learning PettingZoo Environments
- Learning Docker
- Learning PyTorch Geometric
- Learning Electron
- Learning C++
- Learning Streamlit
- Learning FFT
- Learning GPT with Karpathy
- Learning Pogema
- Learning Python
- Learning Seed in Python
- Learning Python Tricks
- Learning cProfile
- Learning Threading in Python
- Learning to save files in Python
- Learning Datetime
- Learning Async IO
- Learning Python Decorators
- Learning f-strings
- Learning Dataclasses
- Learning Type-Checking in Python
- Learning Magic Methods in Python
- Learning Collections
- Learning @property
- Learning Regexes
- Learning Matplotlib Animation
- Learning HTML
- Learning CSS
- Learning CSS Grid
- Learning Flexbox
- Learning CSS Animation
- Learning Flask
- Learning React
- Learning Django
- Learning AJAX
- Learning Async JS and Fetch API
- Learning Bootstrap
- Learning To Deploy Flask App To Heroku
- Learning To Deploy Flask App With DB To Heroku
- Learning Dash (source 1)
- Learning React + Flask + MongpDB
- Learning MongoDB
- DL Course 1: Image Classification with HOG and SVM Techniques
- DL Course 2: TensorFlow and Keras
- Air Hockey Simulation - RL - Tabular Q-learning and Sarsa
- Fuzzy Logic Presentation
- DRL Course 1: MDP, Value Iteration, Policy Iteration
- DRL Course 2: SARSA, Q-learning
- DRL Course 3: DQN, PyTorch
- Greedy Algorithm, Construction Heuristic, Simulated Annealing, Local Search, Genetic Algorithm
- CSP problems
- Stanford cs321n
- Multi-variate Statistics Course (2022)
- Lectures and Assignments
- Flask Begginig
- Jinja2
- Flask Requests
- MySQL
- Blueprints
- Learning Fetch, Requests,
asyncio
in Flask and Plotting Graphs in Flask - REST API
- Bootstrap
- (former repo of the course)
- WEB Course: Flask Skeleton Project (By Barak Pinchovski)
- WEB Course: Lectures - 2021 A
- WEB Course: Lectures - 2021 B
- WEB Course: Lectures - 2022 A - group 1
- WEB Course: Lectures - 2022 A - group 2
- WEB Course: Lectures - 2022 B
- WEB Course: Lectures - 2023 A - group 1
- WEB Course: Lectures - 2023 A - group 2
- WEB Course: Lectures - 2023 B
- WEB Course: Lectures - 2024 A - group 1
- WEB Course: Lectures - 2024 A - group 2
- WEB Course: Lectures - 2024 B
- My Portfolio
- Porftrofilo Template
- Forked: Smart Home Simulator - FinalProject
- Forked: Smart Home Simulator - FinalProjectWrapping
Usefull Links π Mastering Markdown link Emoji Cheat Sheet link Theme of the Page link GitHub Pages Site link GitHub Pages Docs link Budges in GitHub README link
- Toy envs - OpenAI Gym
- SOTA RL envs - minigrid
- SOTA MARL envs - Derkβs Gym, VMAS, MAgent2
- SOTA RL and MARL envs - Unity ML-Agents Toolkit
- SOTA Baselines3
- OpenAI Gym / PettingZoo / MiniGrid
- CityFlow
- MiniHack
- RWARE
- Neural MMO 2.0 (Prev: 1.0)
- Flatland
- highway-env
- Deep RTS
- Gym-ΞΌRTS (pronounced "gym-micro-RTS")
- Nocturne
- CyberBattleSim
- AI Economist - An Economic Simulation Framework
- Neural MMO
- Bigpig4396/Multi-Agent-Reinforcement-Learning-Environment
- PyBullet (no)
- Pogema
- Duckietown Gym Env
- Automatic Parallel Parking: Path Planning, Path Tracking & Control
- PythonRobotics
- Jumanji
- PGX
- FinRL: Financial Reinforcement Learning
- RLCard: A Toolkit for Reinforcement Learning in Card Games
- JaxMARL
- Airlift Challenge v2.0
- overcooked_ai
- gym-cooking
- Aerial Gym Simulator
- Google Research Football
- Griddly
- IMP-MARL: a Suite of Environments for Large-scale Infrastructure Management Planning via MARL
- LBF
- MATE: the Multi-Agent Tracking Environment
- Melting Pot
- Multi-Car Racing Gym Environment
- Nocturne - partially observed, driving simulator