SAIL—Software system for learning AI algorithms

Abstract

Artificial intelligence (AI) comprises a large spectrum of groups of algorithms: heuristic algorithms for search and planning, formal methods for representation of knowledge and reasoning, algorithms for machine learning and many more. Since these algorithms are complex, there is a need for a system which would enable their application both in everyday work and education processes. This paper describes a software system for learning AI algorithms called SAIL (Software System for AI Learning), which can be used both on computers and mobile devices. The paper gives examples of lab exercises and self‐study tasks that through graphic representation and detailed procedures help students master this area. Students can enter their examples into the system and obtain correct solutions for those examples. At any point when an example is simulated, a student can proceed to the next step or go back to the previous one, save the current simulation as a file, or print the detailed procedure as a task solution. SAIL helps lecturers go through the syllabus more efficiently and improve class material, while at the same time it helps students get a better grasp of implemented algorithms. SAIL can also benefit software engineers, who can select and simulate an adequate algorithm to solve a specific problem. The results of the SAIL system are verified within the AI introductory course at the School of Electrical Engineering University of Belgrade and they are presented in this paper.

Application of Unsupervised Migration Method Based on Deep Learning Model in Basketball Training

“…Currently, it appears that most data sets are comparable, allowing migration learning to be broadly applied across a wide range of machine learning domains. Usually, convolutional neural networks are not trained from scratch since, even with strong hardware performance, training the model takes a long time as the dataset size increases [ 5 ]. Thus, the issue of lengthy training is avoided.…”

A unified framework to experiment with algorithm optimality and efficiency

“…Other technical approaches exist to support the instruction of optimization algorithms. The most frequent approach is the simulation of a selected set of specific algorithms [5,12,16], but animation systems for user algorithms [4] also are available. However, the educational goal of both classes of tools is different from ours.…”

Online Data Migration Model and ID3 Algorithm in Sports Competition Action Data Mining Application

“…When the sample m x is read, the positive and negative ratios in the sample set at the previous m moments will change, which will inevitably cause the final classification model to have a learning bias for some samples. To reduce the impact of this change, the current ratio of positive and negative samples and the ratio of the loss of the current sample to the total loss are multiplied together and denoted as D m , as in equation (5). Where H…”

“…At present, it seems that such similarity exists in the vast majority of data, which allows migration learning to be widely applied to many areas of machine learning. Usually, the training of convolutional neural networks does not start from scratch, because as the size of the dataset increases, the time required to train the model is still long even with good hardware performance [5]. This avoids the problem of a long training time.…”