36 skills found · Page 1 of 2
EMI-Group / EvorlEvoRL 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.
MattKleinsmith / PbtPopulation Based Training (in PyTorch with sqlite3). Status: Unsupported
facebookresearch / How To AutorlPlug-and-play hydra sweepers for the EA-based multifidelity method DEHB and several population-based training variations, all proven to efficiently tune RL hyperparameters.
yyzpiero / EVO PopulationBasedTrainingPopulation-Based Training (PBT) for Reinforcement Learning using Message Passing Interface (MPI)
voiler / PopulationBasedTrainingA simple PyTorch implementation of Population Based Training of Neural Networks.
instadeepai / FastpbrlVectorization techniques for fast population-based training.
angusfung / Population Based TrainingReproducing results from DeepMind's paper on Population Based Training of Neural Networks.
fediazgon / Pbt KerasPopulation Based Training of Neural Network implemented in Keras
xingchenwan / Bgpbt[AutoML'22] Bayesian Generational Population-based Training (BG-PBT)
jjccero / PbrlA Population Based Reinforcement Learning Library based on PyTorch
ruizhaogit / Maximum Entropy Population Based TrainingMaximum Entropy Population Based Training for Zero-Shot Human-AI Coordination
bkj / PbtPopulation Based Training, Figure 2
chrischute / PbtPopulation-Based Training in Python
Jackory / RPBT(AAAI24 oral) Implementation of RPPO(Risk-sensitive PPO) and RPBT(Population-based self-play with RPPO)
elsheikh21 / Population Based Training Of NNsApplying PBT optimization technique to different domains
angusfung / Pbt GanApplying Population Based Training on Generative Adversarial Networks.
ArkadiyD / MO PBTAn algorithm for multi-objective hyperparameter optimization: Multi-Objective Population Based Training (MO-PBT), ICML 2023
HumanCompatibleAI / Better Adversarial DefensesTraining in bursts for defending against adversarial policies
cyk1337 / Population Based TrainingPopulation-Based Training (PBT) implementation on ddpg
Kojey / MSc Whistler Waves DetectorLightning strokes create powerful electromagnetic pulses that result in Very Low Frequency (VLF) waves propagating along the magnetic field lines of the earth. Due to the dipole shape of the geomagnetic field, these waves travel upward from the stroke location out through portions of the plasmasphere and back to the Earth’s surface at the field line foot point in the opposite hemisphere. VLF antenna receivers set up at various high and middle latitude locations can detect whistler waves generated by these lightning strokes. The propagation time delay of these waves is dependent on the plasma density along the propagation path. This enables the use of whistler wave observations for characterising the plasmasphere in terms of particle number and energy density. The dynamics of energetic particle populations in the plasmasphere are an important factor in characterising the risk to spacecraft in orbit around Earth. Annual global lightning flash rates are on the order of 45 flash/s [5]. The resulting high occurrence rate of whistler events makes it impossible to identify and characterise them in a reasonable time. Therefore the automatic detection and characterisation of whistlers are valuable to the study of energetic particle dynamics in the plasmasphere and to develop models for operational use. Lichtenberger [1] developed an automatic detector and analyser based on the Appleton-Hartree dispersion relation and experimental models of particle density distribution. Recent advances in artificial neural network-based image processing methods for example, convolutional networks [6] may be able to provide an alternative method for the automatic identification and characterisation of whistler events in broadband VLF spectra. Model development is based on training a neural-network-based model on a large set of spectrograms with whistler events identified by the nodes of the Automatic Whistler Detection and Analysis Network (AWDAnet [7]). Spectrograms will be presented in the form of images (Figure 1.1) to take advantage of the wide range of image-processing techniques available for this type of object identification.