Swarm Optimization

2016-08-25 00:00:00 +0000

Evolutionary algorithms have been used to solve multi-objective optimization problems of two or three objectives giving results that both converge to the optimal front and are diverse along that front. However, once the number of objectives rises to four and above, the Pareto dominance concept that most of these algorithms are based on loses selection pressure, recombination operations are ineffective as individuals in populations are not close in the problem space and the evaluation of performance measures such as hyper-volume becomes computationally expensive.

To overcome the loss in selection pressure, recent evolutionary algorithms adopt a number of techniques which include modifying the dominance relation to increase selection pressure as done with ε-dominance, α-dominance and dominance area control, using a secondary selection mechanism such as the shift-based density estimator, grid-based fitness metric or knee-point driven analysis to filter individuals and decomposing the so-called many objective problems into sub- problems to simultaneously solve.

The controlling dominance area of solutions, CDAS technique has been found to be effective in generating solutions that converge to the optimal front but leads to a loss in the diversity of the generated solutions. In this paper, we propose an extension to the speed-constrained particle swarm optimizer that uses a relaxed form of dominance, CDAS and the shift-based density estimator as secondary selection mechanism to increase diversity of solutions obtained. Also, the particles in the swarm will have an archive of personal bests from which the selection will be done using the weighted sum approach. Finally, the performance of the algorithm will be compared with state of the art algorithms such as NSGA-III and MOEA/DD using the DTLZ benchmark functions.

Link to project
Link to source

Artificial Intelligence

2016-08-24 00:00:00 +0000

Artificial intelligence problems can be classified as state space problems whose solutions are modelled as a series of steps from an initial configuration (initial state) to a final configuration (goal state). Such state space problems typically have various solutions with different costs associated with each. The aim is then to find the series of steps with the minimal cost in the shortest amount of time. Algorithms considered in this category are uninformed search algorithms (breadth first search and depth first search) and informed search algorithms (a-star search, simulated annealing and tabu search).

Similarly, artificial intelligence problems can be solved by modelling the evolutionary process or the learning process of the brain. Evolution is a process that constantly recombines a population of solutions to create a new generation of solutions by weeding out weaker solutions and generating new solutions by combining properties of the fitter solutions. Genetic algorithm will be considered.

Furthermore, algorithms can be developed that can be trained to recognize patterns from a set of training and use that knowledge to classify data not available in the training set. This is demonstrated using a multilayer perceptron. These algorithms are implemented in Java using object-oriented design principles. It is observed that applying heuristics to a search space significantly increases the quality of results both in terms of space and time. Also, even simplistic models of natural processes can be used to find acceptable solutions to otherwise complex problems as demonstrated using genetic algorithms and artificial neural networks.

Link to project
Link to source

Auto-Store

2016-08-23 00:00:00 +0000

Auto-Store is an online database of auto-mechanic stores along with details such as auto parts, prices and locations of those stores. It also allows searching for specific items in the database with various filters such as location, auto parts etc. The application is hosted using a cloud service called Heroku that has support for composer, git, laravel and easy database integration. While this service is paid, they do have a free plan that hosts the app for free for up to 18 hours in any 24-hour period. This means a visit to the link might display a page not found error hosted if it is in the inactive period of the 24 hours.

The web application was built using Laravel framework with a MySQL database for persistence. Bootstrap was used to built the user interface.

Click here to visit the web app.
Link to source

Topsavr

2016-08-22 00:00:00 +0000

TopSavr is a Java-based impulsive savings solution built using Spring boot, Hibernate and Oracle. The thymeleaf template engine was used to design the user interface.

This application is still under development and is a proprietary solution hence very little can be discussed on it.