exilefaker
Developer, artist and (academic) philosopher with interests in machine learning (unsupervised learning, generative models, NLP) and Bayesian methods
New York, NY
Pinned Repositories
biggan
capsnet-experiments
Experiments with variations on the CapsNet architecture using predictive coding and multiple capsule layers
char-rnn
Multi-layer Recurrent Neural Networks (LSTM, GRU, RNN) for character-level language models in Torch
cscg-rr
Research on relative representations for discrete state spaces (using CSCG/CHMMs as a testbed)
deepAI_paper
ekebot_rnn
policy-embeddings
Experiments using policy embeddings and clustering to efficiently explore policy spaces for POMDPs
pymdp
A Python implementation of active inference for Markov Decision Processes
python-UL
Command line tools for unsupervised learning in python/numPy
transformer
Pytorch implementation of the Transformer architecture
exilefaker's Repositories
exilefaker/capsnet-experiments
Experiments with variations on the CapsNet architecture using predictive coding and multiple capsule layers
exilefaker/cscg-rr
Research on relative representations for discrete state spaces (using CSCG/CHMMs as a testbed)
exilefaker/transformer
Pytorch implementation of the Transformer architecture
exilefaker/biggan
exilefaker/char-rnn
Multi-layer Recurrent Neural Networks (LSTM, GRU, RNN) for character-level language models in Torch
exilefaker/deepAI_paper
exilefaker/ekebot_rnn
exilefaker/policy-embeddings
Experiments using policy embeddings and clustering to efficiently explore policy spaces for POMDPs
exilefaker/pymdp
A Python implementation of active inference for Markov Decision Processes
exilefaker/python-UL
Command line tools for unsupervised learning in python/numPy
exilefaker/pytorch-pretrained-BigGAN
🦋A PyTorch implementation of BigGAN with pretrained weights and conversion scripts.
exilefaker/wake-sleep-modifications
Implementations of some modified versions of the wake-sleep algorithm for training a directed neural network with independent bernoulli priors on MNIST.