Code for our NeurIPS 2021 paper: Sparsely Changing Latent States for Prediction and Planning in Partially Observable Domains

Related tags

Deep Learning GateL0RD

Overview

GateL0RD

This is a lightweight PyTorch implementation of GateL0RD, our RNN presented in "Sparsely Changing Latent States for Prediction and Planning in Partially Observable Domains".

We provide two variants of GateL0RD: GateL0RD can be used like a regular PyTorch RNN, whereas GateL0RDCell can be used like a PyTorch RNNCell. To install put gatel0rd.py into your working directory.

Generic example using GateL0RD:

from gatel0rd import GateL0RD
#...
model = GateL0RD(input_size=input_dim, output_size=output_dim, hidden_size=args.latent_dim, reg_lambda=args.lambda)
optimizer = optim.Adam(model.parameters(), lr=args.lr) # optimizer of your choice
# ...
for X,Y in training_data:
    Y_hat, H, Theta = model.forward(X)
    optimizer.zero_grad()
    loss_task = F.mse_loss(Y_hat, Y) # loss of your choice
    loss = model.loss(loss_task, Theta)
    loss.backward()
    optimizer.step()
#...

A repository containing all experiments of the paper, including examples on how to use a GateL0RDCell, can be found here.

BESS: Balanced Evolutionary Semi-Stacking for Disease Detection via Partially Labeled Imbalanced Tongue Data

Balanced-Evolutionary-Semi-Stacking Code for the paper ''BESS: Balanced Evolutionary Semi-Stacking for Disease Detection via Partially Labeled Imbalan

0 Jan 16, 2022

I have created this Virtual Paint Program, in this you can paint(draw) on your screen using hand gestures, created in Python-3 using OpenCV and Mediapipe library. Gestures :- Index Finger for drawing and Index+Middle Finger for changing position and objects.

Virtual-Paint I have created this Virtual Paint Program, in this you can paint(draw) on your screen using hand gestures, created in Python-3. Gestures

6 Sep 22, 2021

Comments

Allow None as hidden state input

The dimensionality of h_tminus1 is checked before it is checked for None, resulting in an error if h_tminus1 is actually set to None. This pull request reverses the order, thus allowing None for h_tminus1, which is then replaced by a zero tensor.

opened by fedorsc 0

Code for our NeurIPS 2021 paper: Sparsely Changing Latent States for Prediction and Planning in Partially Observable Domains

Related tags

Overview

GateL0RD

You might also like...

BESS: Balanced Evolutionary Semi-Stacking for Disease Detection via Partially Labeled Imbalanced Tongue Data

I have created this Virtual Paint Program, in this you can paint(draw) on your screen using hand gestures, created in Python-3 using OpenCV and Mediapipe library. Gestures :- Index Finger for drawing and Index+Middle Finger for changing position and objects.

SLAMP: Stochastic Latent Appearance and Motion Prediction

Code for Parameter Prediction for Unseen Deep Architectures (NeurIPS 2021)

PyTorch implementation for our NeurIPS 2021 Spotlight paper "Long Short-Term Transformer for Online Action Detection".

Diverse Image Captioning with Context-Object Split Latent Spaces (NeurIPS 2020)

Changing the Mind of Transformers for Topically-Controllable Language Generation

Interpretation of T cell states using reference single-cell atlases

😊 Python module for face feature changing

Comments

Allow None as hidden state input

Owner

Autonomous Learning Group

Minimal PyTorch implementation of Generative Latent Optimization from the paper "Optimizing the Latent Space of Generative Networks"

Related resources for our EMNLP 2021 paper

PyTorch implementation for Partially View-aligned Representation Learning with Noise-robust Contrastive Loss (CVPR 2021)

(CVPR 2022) A minimalistic mapless end-to-end stack for joint perception, prediction, planning and control for self driving.

Code for our NeurIPS 2021 paper Mining the Benefits of Two-stage and One-stage HOI Detection

[CVPR2021] DoDNet: Learning to segment multi-organ and tumors from multiple partially labeled datasets

Code for our paper Aspect Sentiment Quad Prediction as Paraphrase Generation in EMNLP 2021.

Code to reproduce the experiments from our NeurIPS 2021 paper " The Limitations of Large Width in Neural Networks: A Deep Gaussian Process Perspective"

Code for our NeurIPS 2021 paper 'Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation'

CowHerd is a partially-observed reinforcement learning environment