'Aligned mixture of latent dynamical systems' (amLDS) for stimulus decoding probabilistic manifold alignment across animals. P. Herrero-Vidal et al. NeurIPS 2021 code.

Last update: Jul 12, 2022

Related tags

Deep Learning amLDS

Overview

Across-animal odor decoding by probabilistic manifold alignment (NeurIPS 2021)

This repository is the official implementation of aligned mixture of latent dynamical systems (amLDS) published at NeurIPS 2021.

amLDS is a probabilistic method to align neural responses and efficiently decode stimuli across animals. It learns independent mappings of different recordings into a shared latent manifold, where stimulus-evoked dynamics are similar (identical) across animals but distint across stimuli allowing for accurate stimulus decoding.

A full description of the algorithm can be found in the preprint.

Requirements

numpy == 1.16.2
matplotlib == 3.0.3
seaborn == 0.9.0
sklearn == 0.20.3
scipy == 1.2.1
python == 3.7.3+

Usage

To get started, run the example notebook 'amLDS_example'. This notebook contains an example on the use of amLDS on synthetic data. It shows how to perform parameter learning, inference and stimulus decoding; as well as latent dimensionality estimation.

To explore other properties and capabilities of amLDS check the 'amLDS_mixturesConcentration' notebook or run the performance script as python3 'amLDS_Performance_DataDemands_ModelComparison.py'.

Copyrights and license

This code has been released under the GNU AGPLv3 license. For the usage of modification of any of this repository content cite:

Pedro Herrero-Vidal, Dmitry Rinberg, Cristina Savin, "Across-animal odor decoding by probabilistic manifold alignment". bioRxiv, 2021

'Aligned mixture of latent dynamical systems' (amLDS) for stimulus decoding probabilistic manifold alignment across animals. P. Herrero-Vidal et al. NeurIPS 2021 code.

Related tags

Overview

Across-animal odor decoding by probabilistic manifold alignment (NeurIPS 2021)

Requirements

Usage

Copyrights and license

Owner

Pedro Herrero-Vidal

A python/pytorch utility library

A PyTorch implementation of the paper "Semantic Image Synthesis via Adversarial Learning" in ICCV 2017

Quantized models with python

TUPÃ was developed to analyze electric field properties in molecular simulations

A python interface for training Reinforcement Learning bots to battle on pokemon showdown

Let Python optimize the best stop loss and take profits for your TradingView strategy.

StarGAN-ZSVC: Unofficial PyTorch Implementation

Source code for our paper "Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures"

Instant Real-Time Example-Based Style Transfer to Facial Videos

Skyformer: Remodel Self-Attention with Gaussian Kernel and Nystr\"om Method (NeurIPS 2021)

Sample Code for "Pessimism Meets Invariance: Provably Efficient Offline Mean-Field Multi-Agent RL"

Neurons Dataset API - The official dataloader and visualization tools for Neurons Datasets.

Datasets and source code for our paper Webly Supervised Fine-Grained Recognition: Benchmark Datasets and An Approach

imbalanced-DL: Deep Imbalanced Learning in Python

QAT(quantize aware training) for classification with MQBench

Python based Advanced AI Assistant

LaBERT - A length-controllable and non-autoregressive image captioning model.

BirdCLEF 2021 - Birdcall Identification 4th place solution

Codecov coverage standard for Python

Episodic Transformer (E.T.) is a novel attention-based architecture for vision-and-language navigation. E.T. is based on a multimodal transformer that encodes language inputs and the full episode history of visual observations and actions.