AIbase
EN
Home
Testpushblock
T

Testpushblock

Developed by rebolforces
A deep reinforcement learning agent trained using the PPO algorithm for Unity's PushBlock game environment
Molecular Model
TensorBoard
#Unity Game AI#Deep Reinforcement Learning#Sokoban Game
Downloads 30
Release Time : 8/20/2022

Model Overview

This model is trained using the Unity ML-Agents framework, specifically designed to solve object-pushing tasks in the PushBlock environment, showcasing the application of reinforcement learning in game AI

Model Features

Based on PPO Algorithm
Trained using Proximal Policy Optimization, a stable reinforcement learning algorithm
Unity Environment Integration
Designed specifically for Unity ML-Agents' PushBlock environment and can be directly deployed in Unity
Continuous Training Support
Supports continued model training via the mlagents-learn command

Model Capabilities

Decision-making in game environments
Solving object-pushing tasks
Executing reinforcement learning policies

Use Cases

Game AI
PushBlock Game AI
Automatically completes object-pushing tasks in the PushBlock environment
Agent performance can be observed in the browser demo
Reinforcement Learning Research
PPO Algorithm Validation
Serves as an implementation case of the PPO algorithm in a simple 3D environment
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase