Population Based Training (in PyTorch with sqlite3). Status: Unsupported

EvoRL is a fully GPU-accelerated framework for Evolutionary Reinforcement Learning, implemented with JAX. It supports Reinforcement Learning (RL), Evolutionary Computation (EC), Evolution-guided Reinforcement Learning (ERL), AutoRL, and seamless integration with GPU-optimized simulation environments.

A simple PyTorch implementation of Population Based Training of Neural Networks.

Reproducing results from DeepMind's paper on Population Based Training of Neural Networks.

Vectorization techniques for fast population-based training.

Population-Based Training (PBT) for Reinforcement Learning using Message Passing Interface (MPI)

A Population Based Reinforcement Learning Library based on PyTorch

Population Based Training, Figure 2

(AAAI24 oral) Implementation of RPPO(Risk-sensitive PPO) and RPBT(Population-based self-play with RPPO)

Training in bursts for defending against adversarial policies

Applying Population Based Training on Generative Adversarial Networks.

Population-Based Training (PBT) implementation on ddpg

My attempt to reproduce a water down version of PBT (Population based training) for MARL (Multi-agent reinforcement learning) using DDPPO (Decentralized & distributed proximal policy optimization) from ray[rllib].

Generating Evolutionary Opponents as a Reinforcement Guided Exploration Solution

Jupyter notebooks to play around with population based training, as described in https://arxiv.org/abs/1711.09846

🥕 Mastering flappy bird with machine learning (neural networks, neuro-evolution)

Curriculum training a Differentiable Neural Computer using Population Based Training

Optimize for topology, hyperparameters, and weights of Neural Nets in single joint training process using Augmented Population Based Training

Article 1 code with bonus Population Based Training support

Implementation of the Google DeepMind paper introducing population-based training, except applied to simulated annealing instead of neural networks.