{"type":"ScholarlyArticle","title":"How to Use Limited Memory in Heuristic Search","authors":[{"name":"Hermann Kaindl","slug":"hermann-kaindl","given":"Hermann","family":"Kaindl","affiliation":"Siemens AG Osterreich Geusaugasse 17 A-1030 Wien , Austria"},{"name":"Gerhard Kainz","slug":"gerhard-kainz","given":"Gerhard","family":"Kainz","affiliation":""},{"name":"Angelika Leeb","slug":"angelika-leeb","given":"Angelika","family":"Leeb","affiliation":""},{"name":"Harald Smetana","slug":"harald-smetana","given":"Harald","family":"Smetana","affiliation":""}],"abstract":"Traditional best-first search for optimal solutions quickly runs out of space even for problem instances of moderate size, and linearspace search has unnecessarily long running times since it cannot make use of available memory. For using available memory effectively, we developed a new generic approach to heuristic search. It integrates various strategies and includes ideas from bidirectional search . Due to insights into differe nt utilizations of available memory, it allows the search to use limited me mory effectively. Instantiations of this approach for two different benchmark domains showed exce ll e nt results that are statistically significant improve ments over previously re ported results: for finding optima l solutions in the 15-Puzzle we achieved the fastest searches of all those using the Manhattan distance heuristic as the only knowledge source, and for a scheduling domain our approach can solve much more difficult problems than the b es t competitor. The most important lessons we learned from the experiments are first, that also in domains with symmetric graph topology selecting the right search direction can be very important , and second, that memory canunder certain conditions- be used much more e ffect iv e ly than by traditional best-first search.","doi":null,"url":"https://ijcai.org/proceedings/1995/31","pdf":"https://ijcai.org/proceedings/1995/0031.pdf","markdown":"https://ijcai.org/proceedings/1995/31.md","bibtex":"https://ijcai.org/proceedings/1995/bibtex/31","volume":{"slug":"1995","title":"Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 1995","url":"https://ijcai.org/proceedings/1995/","year":1995,"location":"Montréal, Québec, Canada","dates":"","editors":"Chris S. Mellish","isbn":"","publisher":"International Joint Conferences on Artificial Intelligence","doi_prefix":""},"section":"Automated Reasoning","track":"Automated Reasoning","number":31,"pages":{"first":236,"last":241},"keywords":[],"videos":[],"references":[{"position":1,"raw":"[Chakrabarti et al., 1989] P.P. Chakrabarti, S. Ghose, A. Acharya, and S. C. DeSarkar . Heuristic search in restri cted memory. Artificial Intelligence , 41(2):197- 221, 1989.","title":"Heuristic search in restri cted memory","authors":"P P Chakrabarti, S Ghose, A Acharya, S C Desarkar","year":"1989","venue":"Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[de Champeaux , 1983] D. de Champeaux. Bidirectional heuristic search again. J. ACM, 30:22-32 , 1983 .","title":"Bidirectional heuristic search again","authors":"D De Champeaux","year":"1983","venue":"J. ACM","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Dechter and Pearl, 1985] R. Dechter and J . Pearl. Gen- eralized best-first strategies and the optimality of A*. J. ACM, 32(3):505-536, 1985.","title":"Generalized best-first strategies and the optimality of A","authors":"R Dechter, J Pearl","year":"1985","venue":"J. ACM","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Dillenburg and Nelson, 1994] J .F. Dillenburg and P.C . Nelson . Perimeter search. Artificial Int elligence, 65:165-178 , 1994.","title":"Perimeter search","authors":"J F Dillenburg, P C Nelson","year":"1994","venue":"Artificial Int elligence","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Ghosh et a/. , 1994] S. Ghosh , A. Mahanti , and D.S. Nau. ITS: an efficient limited-memory heuristic tree search algorithm. In Proc. AAAI-94, pages 1353-1358, 1994.","title":"ITS: an efficient limited-memory heuristic tree search algorithm","authors":"S Ghosh, A Mahanti, D S Nau","year":"1994","venue":"Proc. AAAI-94","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Hart et al., 1968] P.E . Hart, N.J . Nilsson, and B. Raphael. A formal basis for the heuristic determi- nation of minimum cost paths. IEEE Transactions on Systems Science and Cybernetics (SSG), SSC- 4(2):100-107, 1968.","title":"A formal basis for the heuristic determination of minimum cost paths","authors":"P E Hart, N J Nilsson, B Raphael","year":"1968","venue":"IEEE Transactions on Systems Science and Cybernetics (SSG)","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Kaindl and Khorsand , 1994] H. Kaindl and A. Khorsand. Memory-bounded bidirectional search . In Proc. AAAI-94, pages 1359-1364, 1994.","title":"Memory-bounded bidirectional search","authors":"H Kaindl, A Khorsand","year":"1994","venue":"Proc. AAAI-94","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Kaindl and Smetana, 1994] H. Kaindl and H. Smetana. Experimental comparison of heuristic search algo- rithms. In AAAI-94 Workshop on Experimental Eval- uation of R easoning and Search Methods , pages 11-14, 1994.","title":"Experimental comparison of heuristic search algorithms","authors":"H Kaindl, H Smetana","year":"1994","venue":"AAAI-94 Workshop on Experimental Evaluation of R easoning and Search Methods","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Kaindl et a/., 1994] H. Kaindl, A. Leeb, and H. Smetana. Improvements on linear-space search algo- rithms. In Proc. ECAI-94, pages 155-159, 1994.","title":"Improvements on linear-space search algorithms","authors":"H Kaindl, A Leeb, H Smetana","year":"1994","venue":"Proc. ECAI-94","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Kainz , 1995] G . Kainz . Heuristische Suche mit begrenz- tem Speicherbedarf. Doctoral dissertation , Technische Universitat Wien, 1995. Forthcoming.","title":"Heuristische Suche mit begrenztem Speicherbedarf","authors":"G Kainz","year":"1995","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Koll and Kaindl, 1993] A.L. Koll and H. Kaindl. Bidi- rectional best-first search with bounded error : Sum- mary of results. In Proc. IJCAI-93, pages 217-223 , Chambery, France, 1993.","title":"Bidirectional best-first search with bounded error : Summary of results","authors":"A L Koll, H Kaindl","year":"1993","venue":"Proc. IJCAI-93","doi":null,"url":null,"internal_url":"/proceedings/1993/31"},{"position":12,"raw":"[Korf, 1985] R.E. Korf. Depth-first iterative deepening: An optimal admissible tree search. Artificial Int elli- gence, 27(1):97-109 , 1985.","title":"Depth-first iterative deepening: An optimal admissible tree search","authors":"R E Korf","year":"1985","venue":"Artificial Int elligence","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Korf, 1993] R.E. Korf. Linear-space best-first searc:h. Artificial Intellig ence, 62( 1) :41-78, 1993.","title":"Linear-space best-first searc:h","authors":"R E Korf","year":"1993","venue":"Artificial Intellig ence","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Kwa, 1989] J .B.H . Kwa. BS*: An Admissible Bidirec- tional Staged Heuristic Search Algorithm . Artificial Intelligence, 38(2) :95-109, 1989.","title":"An Admissible Bidirectional Staged Heuristic Search Algorithm","authors":"J B H Kwa, Bs","year":"1989","venue":"Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Lawler and Wood, 1966] E.L. Lawler and D. Wood. Branch-and-bound methods: a survey. Operations R e- search, 14:699-719, 1966.","title":"Branch-and-bound methods: a survey","authors":"E L Lawler, D Wood","year":"1966","venue":"Operations R esearch","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Pohl, 1971] I. Pohl. Bi-directional search. In Ma- chine Intelligence 6, pages 127-140 , Edinburgh, 197 1. Edinburgh University Press.","title":"Bi-directional search","authors":"; I Pohl, Pohl","year":"1971","venue":"Machine Intelligence","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Rao et a/., 1991] V.N. Rao, V. Kumar , and R .E. Korf. Depth-first vs best-first search. In Proc. AAAI-91 , pages 434-440, Anaheim , 1991. Los Altos , CA.: Kauf- mann .","title":"Depth-first vs best-first search","authors":"V N Rao, V Kumar, R E Korf","year":"1991","venue":"Proc. AAAI-91","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Reinefeld and Marsland, 1994] A. Reinefeld and T.A. Marsland . Enhanced iterative-deepening search. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAM!} , 16(12):701-709 , July 1994 .","title":"Enhanced iterative-deepening search","authors":"A Reinefeld, T A Marsland","year":"1994","venue":"IEEE Transactions on Pattern Analysis and Machine Intelligence (PAM!}","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Russell, 1992] S. Russell. Efficient memory-bounded search methods. In Proc. ECAI-92, pages 1-5, Vien na, Austria, 1992 . Chichester: Wiley.","title":"Efficient memory-bounded search methods","authors":"S Russell","year":"1992","venue":"Proc. ECAI-92","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Sen and Bagchi, 1989] A.K. Sen and A. Bagchi. Fast recursive formulations for best-first search that allow controlled use of memory. In Proc. IJCAI-89, pages 297-302, 1989 .","title":"Fast recursive formulations for best-first search that allow controlled use of memory","authors":"A K Sen, A Bagchi","year":"1989","venue":"Proc. IJCAI-89","doi":null,"url":null,"internal_url":"/proceedings/1989/47"},{"position":21,"raw":"[Sen and Bagchi, 1993] A.K. Sen and A. Bagchi . Job se- quencing with quadratic penalties : an A* -based graph search approach. In Proc. CAIA-93, pages 190-196, Orlando, FL, March 1993.","title":"Job sequencing with quadratic penalties : an A* -based graph search approach","authors":"A K Sen, A Bagchi","year":"1993","venue":"Proc. CAIA-93","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Taylor and Korf, 1993] L.A. Taylor and R.E. Korf. Pruning duplicate nodes in depth-first search. In Proc. AAAI-93, pages 756-761, Washington, D.C., 1993. Los Altos, CA .: Kaufmann .","title":"Pruning duplicate nodes in depth-first search","authors":"L A Taylor, R E Korf","year":"1993","venue":"Proc. AAAI-93","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Townsend , 1978] W . Townsend. The single machine problem with quadratic penalty fun ction of comple- tion times: a branch-and-bound solution. Manage- ment Sci ence , 24(5) :530-534 , 1978 .","title":"The single machine problem with quadratic penalty fun ction of completion times: a branch-and-bound solution","authors":"W Townsend","year":"1978","venue":"Management Sci ence","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zhang and Korf, 1993] W. Zhang and R.E. Korf. Depth-first vs. best-first search : new results. In Proc. AAAI-93, pages 769-775, Washington, D.C., 1993. Los Altos, CA.: Kaufmann .","title":"Depth-first vs. best-first search : new results","authors":"W Zhang, R E Korf","year":"1993","venue":"Proc. AAAI-93","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}