ThomasMiconi

AMAbot

Code for training AMAbots ("Ask Me Anything" bots) that answer questions in the style of a specific person

BasicV1

A simple model of V1 receptive field learning in 25 lines of Python

BiologicallyPlausibleLearningRNN

Code for "Biologically plausible learning in recurrent neural networks", Miconi et al. eLife 2017

BOHP_RNN

Backprop training of recurrent neural networks with Hebbian plastic connections

DiffRNN

RNN with differentiable structure (number of neurons)

HebbianCNNPyTorch

Automatic Hebbian learning in multi-layer convolutional networks with PyTorch, by expressing Hebbian plasticity rules as gradients

LearningToLearnBOHP

Backpropagation training of neural networks with Hebbian plastic connections

LearningToLearnCogTasks

Code for the ICML 2023 paper "Learning to acquire novel cognitive tasks with evolution, plasticity and meta-meta-learning"

Meta-Task-Generator

Automatically generate simple meta-learning tasks from a very large space

MetaMetaLearning

Evolve plastic networks to be able to automatically acquire novel cognitive (meta-learning) tasks

ThomasMiconi/BiologicallyPlausibleLearningRNN

Code for "Biologically plausible learning in recurrent neural networks", Miconi et al. eLife 2017

ThomasMiconi/HebbianCNNPyTorch

Automatic Hebbian learning in multi-layer convolutional networks with PyTorch, by expressing Hebbian plasticity rules as gradients

ThomasMiconi/LearningToLearnBOHP

Backpropagation training of neural networks with Hebbian plastic connections

ThomasMiconi/DiffRNN

RNN with differentiable structure (number of neurons)

ThomasMiconi/BOHP_RNN

Backprop training of recurrent neural networks with Hebbian plastic connections

ThomasMiconi/Meta-Task-Generator

Automatically generate simple meta-learning tasks from a very large space

ThomasMiconi/LearningToLearnCogTasks

Code for the ICML 2023 paper "Learning to acquire novel cognitive tasks with evolution, plasticity and meta-meta-learning"

ThomasMiconi/MetaMetaLearning

Evolve plastic networks to be able to automatically acquire novel cognitive (meta-learning) tasks

ThomasMiconi/BasicV1

A simple model of V1 receptive field learning in 25 lines of Python

ThomasMiconi/BOHP

Backpropagation of Hebbian Plasticity

ThomasMiconi/GABugs

Evolving reinforcement learning in a foraging agent

ThomasMiconi/nodeapp

ThomasMiconi/V1stdp

Model of developing V1 from Miconi, McKinstry & Edelman, Nat. Comm. 2016

ThomasMiconi/EvoBugs

RNN-controlled agents freely evolving in a 2D world

ThomasMiconi/FairnessPaper

ThomasMiconi/AMAbot

Code for training AMAbots ("Ask Me Anything" bots) that answer questions in the style of a specific person

ThomasMiconi/Blocksworld

ThomasMiconi/demo_app

ThomasMiconi/frequency-encoder

ThomasMiconi/htmresearch

Experimental algorithms. Unsupported.

ThomasMiconi/htmresearch-core

Fork of nupic.core that also contains Numenta's experimental C++ research code. Please see nupic.research for more details.

ThomasMiconi/nupic

Numenta Platform for Intelligent Computing: a brain-inspired machine intelligence platform, and biologically accurate neural network based on cortical learning algorithms.

ThomasMiconi/RNN_previous

Training recurrent neural networks with sparse, delayed rewards.

ThomasMiconi/sample_app

ThomasMiconi/ThomasMiconi.github.io

ThomasMiconi/TransitiveInference

Code for the preprint "An active neural mechanism for relational learning and fast knowledge reassembly"

ThomasMiconi/VisualSearchModel

Code and data for Miconi, Groomes & Kreiman, Cerebral Cortex 2015