Skip to content

Solving generic challenges of artificial intelligence and cognitive systems using Java

Notifications You must be signed in to change notification settings

Pexers/AI-and-cognitive-systems

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 

Repository files navigation

AI & Cognitive Systems

This repository contains a few projects that solve generic challenges of artificial intelligence and cognitive systems using Java. Every module contains some Gradle tests that validate the respective solution.

Module 1 - Neural Networks

Learning the XOR logic function using the Forward Propagation algorithm for neural networks.

Module 2 - Optimization & Genetic algorithms

Some of the challenges included in this module were inspired by Google's OR-Tools problems:

Illustrated below, is the Assignment with Teams of Workers problem, where the goal was to assign every worker to a single task while trying to minimize the total cost as much as possible. Two workers can't be performing the same task.

Module 3 - Autonomous Agent using Reactive, Adaptive and Deliberative layers

The purpose of this module was to create an agent capable of navigating in a space of obstacles and a target to collect. The agent movement is continuous or discrete, depending on the challenge. The first image shows a reactive agent (continuous movement), while the second image shows an adaptive agent (discrete movement). The deliberative layer uses both the adaptive and reactive layer. The results obtained for the deliberative implementation weren't as good as the other two layers.