L ECTURE 27: S WARM I NTELLIGENCE 8 / A NT C OLONY O PTIMIZATION 4 T - PowerPoint PPT Presentation

15-382 C OLLECTIVE I NTELLIGENCE - S19 L ECTURE 27: S WARM I NTELLIGENCE 8 / A NT C OLONY O PTIMIZATION 4 T EACHER : G IANNI A. D I C ARO

2-OPT LOCAL SEARCH 2

3-OPT LOCAL SEARCH 3

LOCAL SEARCH, ITERATED LOCAL SEARCH Check additional slides: local-search.pdf 4

PHEROMONE AND HEURISTIC ARE BOTH IMPORTANT! 5

ANTS + LOCAL SEARCH • As a matter of fact, the best instances of ACO algorithms for (static/centralized) combinatorial problems are those making use of a problem-speci fi c local search daemon procedure (iterative solution modi fi cation) interleaved with ants’ solution construction • It is conjectured that ACO’s ants can provide good starting points for local search. More in general, a construction heuristic can be used to quickly build up a complete solution of good quality, and then a solution modi fi cation procedure can take this solution as a starting point, trying to further improve it by iteratively modifying some of its parts • This hybrid two-phases search can be iterated and can be very e ff ective if each phase can produce a solution which is locally optimal within a di ff erent class of feasible solutions, with the intersection between the two classes being minimal 6

MAX-MIN-AS (1999): LIMITS AND RESTARTS 7

MMAS (1999): LIMITS AND RESTARTS n 8

AS RANK (1999): ELITISM BY RANK 9

ANT-TABU (2001) 10

DYNAMIC, DISTRIBUTED ENVIRONMENTS? • Routing in wired networks, AntNet (1998) • Routing in mobile ad hoc networks (AntHocNet, 2005) • Routing / Foraging in mobile robotic networks Challenges: Distributed, path evaluation, non-stationary, errors, “unpredictable” tra ffi c demands, interference of ants with normal tra ffi c, limited bandwidth, when/where send ants, where to store pheromone? 11

DECISIONS TO TAKE DESIGNING ACO ALGORITHMS… 12

A FEW COMMON DESIGN CHOICES 13

THEORETICAL RESULTS? 14

ACO SUMMARY • Reverse engineering of stigmergic pheromone laying-following in ant colonies • Construction meta-heuristic biased by pheromone + heuristic • Ants: Monte Carlo sampling of solutions • Generalized policy iteration for learning pheromone parameters for decision-making: policy evaluation (sampling of solutions) + policy improvement (pheromone updating) • Di ff erent strategies for sampling and updating • A number of di ff erent heuristic recipes (common in SI, heuristic optimization domains) • In physically distributed problems (e.g., networks, robotics) agents hops from one decision node to another and then have to retrace their path back, if feasible • State of the art performance (when coupled with LS, for centralized problems) • Guaranteed performance: yes, in the probabilistic limit • Applied to a large variety of CO problems • Large number of scienti fi c publications • Applied in the real world: Barilla, Migros, port management, logistics, …. 15