Recency-based memory

As explained in Optimization Algorithms: AI techniques for design, planning, and control problems

A tabu-search memory that records recently visited moves or search attributes and temporarily restricts their reuse.

Recency-based memory is tabu search’s short-term memory: it records recently visited moves or search attributes and temporarily prevents their reuse. Unlike memoryless local search, it gives the search an explicit history, helping avoid immediate revisits and cycles while exploring neighboring solutions. section 6.2.1

A practical tabu-search cycle uses the memory as follows:

• Record the move or attribute used during the current transition. section 6.2.1
• Mark that move tabu for a limited tabu tenure, measured as the number of iterations before reuse is allowed. section 6.2.1 section 6.2.3
• Adjust the tenure when necessary: short tenures can permit repeated moves too quickly, while long tenures can cause stagnation, larger memory requirements, and slower computation. section 6.2.3

Four-city traveling-salesperson search

A tabu structure for a four-city traveling-salesperson problem can store remaining tabu iterations for active moves. As iterations pass, those counts expire, allowing previously restricted moves to re-enter the neighborhood. The same structure can be contrasted with frequency-based memory, which counts how often moves have been used across the full search. section 6.2.1

Figure 6.7 Dynamically controlling the tabu tenure

Use the dynamic-tenure figure to compare intensification near improving solutions with diversification away from deteriorating or stagnant regions. section 6.2.3

How Optimization Algorithms: AI techniques for design, planning, and control problems covers it

How deep it goes

Explains memory structure, tenure trade-offs, dynamic adjustment, and reactive mechanisms, but does not provide implementation code.

Examples

Uses a four-city traveling-salesperson tabu structure with remaining tabu iterations and move-frequency counts.

What it assumes

Assumes familiarity with local search, moves, neighborhoods, and solutions while contrasting explicit memory with memoryless search.

What you can do after

Operate/configure tabu search by choosing, bounding, and dynamically adapting the tabu tenure.

How it connects

Connects recency-based memory to tabu search, local search, frequency-based memory, aspiration criteria, intensification, and diversification.

liveBookmanning.com

Read Optimization Algorithms: AI techniques for design, planning, and control problems on liveBook Get the book at manning.com

Related concepts

• tabu listimplements · A tabu list is a memory structure that records recently explored moves, locations, or solutions to temporarily prevent their reconsideration during a search.
• memorykind of · Memory is a storage resource or mechanism that retains data or computational state for access during computation or later use.
• tabu searchpart of · Tabu search is a memory-guided local search technique that restricts recently used moves or solutions to discourage cycling and escape local optima.
• working memorycontrasts with · Working memory is a limited, temporary store of information available for immediate access or manipulation during an active task or interaction.

This explainer was generated by Manning's knowledge graph from the text of Optimization Algorithms: AI techniques for design, planning, and control problems. The section links point to the parts of the book it draws on.