{"publisher":"Institute of Science and Technology Austria","oa_version":"Published Version","oa":1,"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","date_created":"2020-05-26T14:08:55Z","alternative_title":["ISTA Thesis"],"title":"On the average-case hardness of total search problems","status":"public","related_material":{"record":[{"status":"public","id":"6677","relation":"part_of_dissertation"}]},"article_processing_charge":"No","supervisor":[{"first_name":"Krzysztof Z","last_name":"Pietrzak","id":"3E04A7AA-F248-11E8-B48F-1D18A9856A87","full_name":"Pietrzak, Krzysztof Z","orcid":"0000-0002-9139-1654"}],"month":"05","date_updated":"2023-09-07T13:15:55Z","tmp":{"image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"type":"dissertation","file":[{"file_id":"7897","date_created":"2020-05-26T14:08:13Z","file_size":1622742,"date_updated":"2020-07-14T12:48:04Z","content_type":"application/pdf","file_name":"2020_Thesis_Kamath.pdf","access_level":"open_access","creator":"dernst","relation":"main_file","checksum":"b39e2e1c376f5819b823fb7077491c64"},{"file_id":"7898","file_size":15301529,"date_created":"2020-05-26T14:08:23Z","date_updated":"2020-07-14T12:48:04Z","content_type":"application/x-zip-compressed","access_level":"closed","creator":"dernst","file_name":"Thesis_Kamath.zip","checksum":"8b26ba729c1a85ac6bea775f5d73cdc7","relation":"source_file"}],"day":"25","degree_awarded":"PhD","department":[{"_id":"KrPi"}],"abstract":[{"text":"A search problem lies in the complexity class FNP if a solution to the given instance of the problem can be verified efficiently. The complexity class TFNP consists of all search problems in FNP that are total in the sense that a solution is guaranteed to exist. TFNP contains a host of interesting problems from fields such as algorithmic game theory, computational topology, number theory and combinatorics. Since TFNP is a semantic class, it is unlikely to have a complete problem. Instead, one studies its syntactic subclasses which are defined based on the combinatorial principle used to argue totality. Of particular interest is the subclass PPAD, which contains important problems\r\nlike computing Nash equilibrium for bimatrix games and computational counterparts of several fixed-point theorems as complete. In the thesis, we undertake the study of averagecase hardness of TFNP, and in particular its subclass PPAD.\r\nAlmost nothing was known about average-case hardness of PPAD before a series of recent results showed how to achieve it using a cryptographic primitive called program obfuscation.\r\nHowever, it is currently not known how to construct program obfuscation from standard cryptographic assumptions. Therefore, it is desirable to relax the assumption under which average-case hardness of PPAD can be shown. In the thesis we take a step in this direction. First, we show that assuming the (average-case) hardness of a numbertheoretic\r\nproblem related to factoring of integers, which we call Iterated-Squaring, PPAD is hard-on-average in the random-oracle model. Then we strengthen this result to show that the average-case hardness of PPAD reduces to the (adaptive) soundness of the Fiat-Shamir Transform, a well-known technique used to compile a public-coin interactive protocol into a non-interactive one. As a corollary, we obtain average-case hardness for PPAD in the random-oracle model assuming the worst-case hardness of #SAT. Moreover, the above results can all be strengthened to obtain average-case hardness for the class CLS ⊆ PPAD.\r\nOur main technical contribution is constructing incrementally-verifiable procedures for computing Iterated-Squaring and #SAT. By incrementally-verifiable, we mean that every intermediate state of the computation includes a proof of its correctness, and the proof can be updated and verified in polynomial time. Previous constructions of such procedures relied on strong, non-standard assumptions. Instead, we introduce a technique called recursive proof-merging to obtain the same from weaker assumptions. ","lang":"eng"}],"publication_status":"published","date_published":"2020-05-25T00:00:00Z","year":"2020","project":[{"call_identifier":"FP7","grant_number":"259668","name":"Provable Security for Physical Cryptography","_id":"258C570E-B435-11E9-9278-68D0E5697425"},{"call_identifier":"H2020","_id":"258AA5B2-B435-11E9-9278-68D0E5697425","name":"Teaching Old Crypto New Tricks","grant_number":"682815"}],"ddc":["000"],"file_date_updated":"2020-07-14T12:48:04Z","publication_identifier":{"issn":["2663-337X"]},"has_accepted_license":"1","author":[{"id":"4BD3F30E-F248-11E8-B48F-1D18A9856A87","full_name":"Kamath Hosdurg, Chethan","last_name":"Kamath Hosdurg","first_name":"Chethan"}],"page":"126","doi":"10.15479/AT:ISTA:7896","_id":"7896","language":[{"iso":"eng"}],"citation":{"ista":"Kamath Hosdurg C. 2020. On the average-case hardness of total search problems. Institute of Science and Technology Austria.","apa":"Kamath Hosdurg, C. (2020). <i>On the average-case hardness of total search problems</i>. Institute of Science and Technology Austria. <a href=\"https://doi.org/10.15479/AT:ISTA:7896\">https://doi.org/10.15479/AT:ISTA:7896</a>","chicago":"Kamath Hosdurg, Chethan. “On the Average-Case Hardness of Total Search Problems.” Institute of Science and Technology Austria, 2020. <a href=\"https://doi.org/10.15479/AT:ISTA:7896\">https://doi.org/10.15479/AT:ISTA:7896</a>.","ama":"Kamath Hosdurg C. On the average-case hardness of total search problems. 2020. doi:<a href=\"https://doi.org/10.15479/AT:ISTA:7896\">10.15479/AT:ISTA:7896</a>","short":"C. Kamath Hosdurg, On the Average-Case Hardness of Total Search Problems, Institute of Science and Technology Austria, 2020.","mla":"Kamath Hosdurg, Chethan. <i>On the Average-Case Hardness of Total Search Problems</i>. Institute of Science and Technology Austria, 2020, doi:<a href=\"https://doi.org/10.15479/AT:ISTA:7896\">10.15479/AT:ISTA:7896</a>.","ieee":"C. Kamath Hosdurg, “On the average-case hardness of total search problems,” Institute of Science and Technology Austria, 2020."},"ec_funded":1}