nnaisense/pgpelib nnaisense/pgpelib

  • Stars
    star
    68
  • Rank 457,643 (Top 10 %)
  • Language
    Python
  • License
    BSD 3-Clause "New...
  • Created about 4 years ago
  • Updated almost 4 years ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
nnaisense/pgpelib
github

nnaisense

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

A mini library for Policy Gradients with Parameter-based Exploration, with reference implementation of the ClipUp optimizer (https://arxiv.org/abs/2008.02387) from NNAISENSE.

More Repositories

1

bayesian-flow-networks

This is the official code release for Bayesian Flow Networks.
Python
232
star
2

MAX

Code for reproducing experiments in Model-Based Active Exploration, ICML 2019
Python
77
star
3

2017-learning-to-run

The Winning Solution for the Learning To Run Challenge 2017
Python
60
star
4

conditional-style-transfer

Two-Stage Peer-Regularized Feature Recombination for Arbitrary Image Style Transfer
Python
42
star
5

deep-iterative-surface-normal-estimation

Code repository for our paper: Deep Iterative Surface Normal Estimation
Python
32
star
6

MAGE

Learning Action-Value Gradients in Model-based Policy Optimization
Python
31
star
7

pytorch_sym3eig

Pytorch extension: Batch-wise eigencomputation for symmetric 3x3 matrices
C++
19
star
8

gmpda

Gaussian Mixture Periodicity Detection Algorithm
Python
8
star
9

evotorch-myosuite-starter

Jupyter Notebook
6
star
10

evotorch

EvoTorch is an advanced evolutionary algorithm library built directly on top of PyTorch and developed by the research scientists at NNAISENSE.
Python
4
star
11

mems

Multimeasurement Generative Models
Python
4
star
12

self-recog-as-optim

Source code for experiments presented in 'Self Recognition as Optimisation' at ALIFE 2022
Python
3
star