Centre for Artificial Intelligence Research (CAIR)

deep-rts

A Real-Time-Strategy game for Deep Learning research

fast-tsetlin-machine-with-mnist-demo

A fast Tsetlin Machine implementation employing bit-wise operators, with MNIST demo.

Fire-Detection-Image-Dataset

This dataset contains normal images and images with fire. It is highly unbalanced to reciprocate real world situations. It consists of a variety of scenarios and different fire situations (intensity, luminosity, size, environment etc).

GraphTsetlinMachine

Tsetlin Machine for Logical Learning and Reasoning With Graphs

pyTsetlinMachine

Implements the Tsetlin Machine, Convolutional Tsetlin Machine, Regression Tsetlin Machine, Weighted Tsetlin Machine, and Embedding Tsetlin Machine, with support for continuous features, multigranularity, clause indexing, and literal budget

pyVNC

VNC Client Library for Python

rl

tmu

Implements the Tsetlin Machine, Coalesced Tsetlin Machine, Convolutional Tsetlin Machine, Regression Tsetlin Machine, and Weighted Tsetlin Machine, with support for continuous features, drop clause, Type III Feedback, focused negative sampling, multi-task classifier, autoencoder, literal budget, and one-vs-one multi-class classifier. TMU is written in Python with wrappers for C and CUDA-based clause evaluation and updating.

TsetlinMachine

Code and datasets for the Tsetlin Machine

TsetlinMachineBook

Python code accompanying the book "An Introduction to Tsetlin Machines".

cair/Fire-Detection-Image-Dataset

This dataset contains normal images and images with fire. It is highly unbalanced to reciprocate real world situations. It consists of a variety of scenarios and different fire situations (intensity, luminosity, size, environment etc).

cair/fast-tsetlin-machine-with-mnist-demo

A fast Tsetlin Machine implementation employing bit-wise operators, with MNIST demo.

cair/convolutional-tsetlin-machine-tutorial

Tutorial on the Convolutional Tsetlin Machine

cair/TextUnderstandingTsetlinMachine

Using the Tsetlin Machine to learn human-interpretable rules for high-accuracy text categorization with medical applications

cair/TsetlinMachineBook

Python code accompanying the book "An Introduction to Tsetlin Machines".

cair/FlashRL

cair/fast-tsetlin-machine-in-cuda-with-imdb-demo

A CUDA implementation of the Tsetlin Machine based on bitwise operators

cair/deep_maze

cair/open-tsetlin-machine

Open Source Tsetlin Machine framework

cair/TsetlinMachineC

A C implementation of the Tsetlin Machine

cair/rl

cair/deep-warehouse

A Simulator for complex logistic environments

cair/regression-tsetlin-machine

Implementation of the Regression Tsetlin Machine

cair/Axis_and_Allies

A simple Axis & Allies engine.

cair/ICML-Massively-Parallel-and-Asynchronous-Tsetlin-Machine-Architecture

Code repository for ICML 21 for Paper titled Massively Parallel and Asynchronous Tsetlin Machine Architecture

cair/ikt111

cair/python-fast-tsetlin-machine

Python wrapper for https://github.com/cair/fast-tsetlin-machine-with-mnist-demo

cair/Fire-Scene-Parsing

cair/deep-line-wars-2

cair/Deterministic-Tsetlin-Machine

Due to the high energy consumption and scalability challenges of deep learning, there is a critical need to shift research focus towards dealing with energy consumption constraints. Tsetlin Machines (TMs) are a recent approach to machine learning that has demonstrated significantly reduced energy usage compared to neural networks alike, while performing competitively accuracy-wise on several benchmarks. However, TMs rely heavily on energy-costly random number generation to stochastically guide a team of Tsetlin Automata to a Nash Equilibrium of the TM game. In this paper, we propose a novel finite-state learning automaton that can replace the Tsetlin Automata in TM learning, for increased determinism. The new automaton uses multi-step deterministic state jumps to reinforce sub-patterns. Simultaneously, flipping a coin to skip every d'th state update ensures diversification by randomization. The d-parameter thus allows the degree of randomization to be finely controlled. E.g., d=1 makes every update random and d=infinity makes the automaton completely deterministic. Our empirical results show that, overall, only substantial degrees of determinism reduces accuracy. Energy-wise, random number generation constitutes switching energy consumption of the TM, saving up to 11 mW power for larger datasets with high d values. We can thus use the new d-parameter to trade off accuracy against energy consumption, to facilitate low-energy machine learning.

cair/fire

cair/covid-19-us-dataset

cair/crl-core

cair/crl-deep-line-wars

cair/crl-example

cair/gym-flashrl

cair/hex-ai

Various AIs for the board game hex, including Monte Carlo Tree Search with the Tsetlin Machine

cair/principles_of_ai

cair/ray-deeprts

cair/tsetlinmachinecuda