Optimization Apparatus Escaping Local Solutions via Stored Energy Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In existing optimization methods, such as simulated annealing, when state transitions are repeatedly rejected, the search process gets stuck in local solutions, leading to increased processing time and reduced efficiency in finding the optimum solution, especially at low thermal noise levels.
Innovation Solution
An optimization program and apparatus that iteratively store energy differences and variable values for states where transitions are rejected, allowing selection of a new state based on these stored values to facilitate transitions and avoid prolonged stays in local solutions, thereby enhancing the search efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If simulated annealing is used to solve optimization problems, then the method can find optimal solutions through stochastic search, but the search process gets stuck in local solutions when transitions are repeatedly rejected, leading to increased processing time
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing energy difference information and variable values for potential state transitions before actually needing to escape a local solution. When the search gets stuck, this pre-stored information is readily available to enable rapid transition to a new state without extensive recalculation, thus reducing the time lost while maintaining the ability to find optimal solutions
Solution Approach 2:
The patent introduces an intermediary mechanism in the form of stored energy difference information and variable values that act as a bridge between the current local solution and potential better solutions. This intermediary data structure allows the system to efficiently evaluate and transition to new states without directly computing all possible transitions, mediating between the stochastic search process and the optimization goal while reducing processing time
2Measurement precision
If thermal noise is reduced to improve solution precision, then the search converges faster to local solutions, but the ability to escape local solutions and find global optimum deteriorates
Solution Approach 1:
The patent implements feedback by continuously monitoring the state transition acceptance rate and detecting when the search has become stuck in a local solution (indicated by consecutive rejections). Based on this feedback, the system activates the stored information mechanism to escape the local solution, thus maintaining the ability to find global optima even when thermal noise is reduced for faster convergence and higher solution precision
Solution Approach 2:
The patent applies dynamics by making the search behavior adaptive rather than static. The system dynamically adjusts its behavior based on the current search state: when transitions are being accepted, it continues normal simulated annealing; when transitions are repeatedly rejected indicating a local solution, it switches to using pre-stored information to escape. This dynamic adaptation allows the system to maintain both solution precision and the ability to escape local solutions
Data Source
AI summary
An optimization apparatus of: performing a search for an optimal solution through iteration of processing configured to determine, by using a difference between a first energy value of an evaluation function in a first state and a second energy value of the evaluation function in a second state, whether to accept a state transition from the first state to the second state, and perform the state transition; and in the performing of the search for the state, storing the difference between the first and second energy values, and the respective values of the evaluation function in the second state, in a case where negation of the state transition is consecutively iterated a specific number of times, selecting one state from among the stored second states, based on the differences, and making the state transition from the first state to the selected one state.