Modified Teaching Learning Based Search Optimization Technique
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional variable selection methods, such as Teaching Learning Based Optimization (TLBO), face challenges in ensuring optimal solutions due to reliance on heuristic approaches that may not explore the complete problem space, leading to local optima and irreproducible results, especially when applied to domains like life sciences and education.
Innovation Solution
A modified teaching learning based search optimization technique that performs exhaustive searches on combined teacher and student variable combinations, introduces additional unique variables in teaching and learning phases, and initializes the population with a minimum objective function value to ensure diversity and reproducibility, thereby minimizing local optimization and stagnation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If heuristic methods such as ant colony, particle swarm optimization, or genetic algorithm are used for variable selection, then the computational complexity is reduced compared to exhaustive search, but the solution optimality cannot be guaranteed as these methods fail to explore the complete problem space
Solution Approach 1:
The patent extracts the deterministic exhaustive search mechanism from the heuristic framework and applies it specifically to evaluate variable combinations generated during the optimization process. This allows the system to maintain computational efficiency while ensuring optimality for the evaluated subsets.
Solution Approach 2:
The patent performs exhaustive search on partial variable combinations rather than the entire variable space. By applying exhaustive evaluation to selected subsets of variables (those generated by the heuristic process), the system achieves optimality for the evaluated portion while maintaining overall computational efficiency.
2Ease of operation
If traditional TLBO method is used for variable selection, then the implementation is simple based on teaching and learning phases, but the results are irreproducible and may converge to local optima due to reliance on heuristic approaches
Solution Approach 1:
The patent introduces feedback mechanisms where the results from exhaustive search are used to update and refine the population for subsequent teaching and learning phases. This feedback loop ensures that the deterministic optimal solutions inform the heuristic search, improving reproducibility while maintaining simplicity.
Solution Approach 2:
The patent performs preliminary exhaustive search to identify optimal variable combinations before applying the TLBO heuristic process. This preliminary action ensures that the starting point for the heuristic optimization is based on proven optimal solutions, eliminating irreproducibility issues.
3Reliability
If exhaustive search approach is used to evaluate each possible combination of variables, then the optimal solution is guaranteed, but the computational complexity becomes intractable for large subsets of variables
Solution Approach 1:
The patent segments the variable selection process into multiple stages: initial heuristic generation of variable combinations, followed by exhaustive evaluation of these segmented subsets, and then iterative refinement. This segmentation makes the exhaustive search computationally manageable by applying it to smaller, manageable portions of the total variable space.
Solution Approach 2:
The patent applies exhaustive search action only to partial variable combinations (those generated by the heuristic process) rather than the complete variable space. This partial application of exhaustive search achieves optimality for the evaluated subsets while avoiding the intractable computational complexity of evaluating all possible combinations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods include initializing a trainees population (TP), calculating an objective function (OF) of the TP to identify a trainer. A teaching pool is created using variables of each trainee and the identified trainer, and unique variables are added to obtain an updated teaching pool (UTP). Search is performed on the UTP to obtain 'm' subset of variables and OFs. The OFs of 'm' subset are compared with OFs of the trainer's and each trainee's variable and one of the trainer or each trainee are updated accordingly. An updated learning pool (ULP) is created for selected trainee and the trainees, by adding unique variables to obtain 'n' subset. The OF of 'n' subset are compared with objective functions of selected trainee and the trainees and variables are updated accordingly. These steps are iteratively performed to obtain an optimal subset of variables that is selected for teaching and learning phase.