[{"isi":1,"oa_version":"Preprint","volume":41,"date_updated":"2026-09-13T22:31:00Z","quality_controlled":"1","fulldoi":"https://doi.org/10.1145/3339984","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","month":"10","article_type":"original","day":"01","year":"2019","title":"Non-polynomial worst-case analysis of recursive programs","doi":"10.1145/3339984","publisher":"ACM","_id":"7014","related_material":{"record":[{"relation":"earlier_version","id":"639","status":"public"},{"status":"public","id":"8934","relation":"dissertation_contains"}]},"citation":{"short":"K. Chatterjee, H. Fu, A.K. Goharshady, ACM Transactions on Programming Languages and Systems 41 (2019).","apa":"Chatterjee, K., Fu, H., &#38; Goharshady, A. K. (2019). Non-polynomial worst-case analysis of recursive programs. <i>ACM Transactions on Programming Languages and Systems</i>. ACM. <a href=\"https://doi.org/10.1145/3339984\">https://doi.org/10.1145/3339984</a>","mla":"Chatterjee, Krishnendu, et al. “Non-Polynomial Worst-Case Analysis of Recursive Programs.” <i>ACM Transactions on Programming Languages and Systems</i>, vol. 41, no. 4, 20, ACM, 2019, doi:<a href=\"https://doi.org/10.1145/3339984\">10.1145/3339984</a>.","chicago":"Chatterjee, Krishnendu, Hongfei Fu, and Amir Kafshdar Goharshady. “Non-Polynomial Worst-Case Analysis of Recursive Programs.” <i>ACM Transactions on Programming Languages and Systems</i>. ACM, 2019. <a href=\"https://doi.org/10.1145/3339984\">https://doi.org/10.1145/3339984</a>.","ieee":"K. Chatterjee, H. Fu, and A. K. Goharshady, “Non-polynomial worst-case analysis of recursive programs,” <i>ACM Transactions on Programming Languages and Systems</i>, vol. 41, no. 4. ACM, 2019.","ama":"Chatterjee K, Fu H, Goharshady AK. Non-polynomial worst-case analysis of recursive programs. <i>ACM Transactions on Programming Languages and Systems</i>. 2019;41(4). doi:<a href=\"https://doi.org/10.1145/3339984\">10.1145/3339984</a>","ista":"Chatterjee K, Fu H, Goharshady AK. 2019. Non-polynomial worst-case analysis of recursive programs. ACM Transactions on Programming Languages and Systems. 41(4), 20."},"project":[{"grant_number":"ICT15-003","_id":"25892FC0-B435-11E9-9278-68D0E5697425","name":"Efficient Algorithms for Computer Aided Verification"},{"grant_number":"S 11407_N23","call_identifier":"FWF","_id":"25832EC2-B435-11E9-9278-68D0E5697425","name":"Rigorous Systems Engineering"},{"name":"Quantitative Graph Games: Theory and Applications","_id":"2581B60A-B435-11E9-9278-68D0E5697425","call_identifier":"FP7","grant_number":"279307"},{"name":"Quantitative Analysis of Probabilistic Systems with a focus on Crypto-Currencies","_id":"267066CE-B435-11E9-9278-68D0E5697425"},{"_id":"266EEEC0-B435-11E9-9278-68D0E5697425","name":"Quantitative Game-theoretic Analysis of Blockchain Applications and Smart Contracts"}],"publication_status":"published","date_created":"2019-11-13T08:33:43Z","article_number":"20","type":"journal_article","issue":"4","author":[{"full_name":"Chatterjee, Krishnendu","orcid":"0000-0002-4561-241X","first_name":"Krishnendu","last_name":"Chatterjee","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Fu, Hongfei","first_name":"Hongfei","last_name":"Fu"},{"id":"391365CE-F248-11E8-B48F-1D18A9856A87","last_name":"Goharshady","first_name":"Amir Kafshdar","full_name":"Goharshady, Amir Kafshdar","orcid":"0000-0003-1702-6584"}],"main_file_link":[{"url":"https://arxiv.org/abs/1705.00317","open_access":"1"}],"arxiv":1,"intvolume":"        41","abstract":[{"text":"We study the problem of developing efficient approaches for proving\r\nworst-case bounds of non-deterministic recursive programs. Ranking functions\r\nare sound and complete for proving termination and worst-case bounds of\r\nnonrecursive programs. First, we apply ranking functions to recursion,\r\nresulting in measure functions. We show that measure functions provide a sound\r\nand complete approach to prove worst-case bounds of non-deterministic recursive\r\nprograms. Our second contribution is the synthesis of measure functions in\r\nnonpolynomial forms. We show that non-polynomial measure functions with\r\nlogarithm and exponentiation can be synthesized through abstraction of\r\nlogarithmic or exponentiation terms, Farkas' Lemma, and Handelman's Theorem\r\nusing linear programming. While previous methods obtain worst-case polynomial\r\nbounds, our approach can synthesize bounds of the form $\\mathcal{O}(n\\log n)$\r\nas well as $\\mathcal{O}(n^r)$ where $r$ is not an integer. We present\r\nexperimental results to demonstrate that our approach can obtain efficiently\r\nworst-case bounds of classical recursive algorithms such as (i) Merge-Sort, the\r\ndivide-and-conquer algorithm for the Closest-Pair problem, where we obtain\r\n$\\mathcal{O}(n \\log n)$ worst-case bound, and (ii) Karatsuba's algorithm for\r\npolynomial multiplication and Strassen's algorithm for matrix multiplication,\r\nwhere we obtain $\\mathcal{O}(n^r)$ bound such that $r$ is not an integer and\r\nclose to the best-known bounds for the respective algorithms.","lang":"eng"}],"date_published":"2019-10-01T00:00:00Z","department":[{"_id":"KrCh"}],"oa":1,"language":[{"iso":"eng"}],"external_id":{"arxiv":["1705.00317"],"isi":["000564108400001"]},"ec_funded":1,"article_processing_charge":"No","publication":"ACM Transactions on Programming Languages and Systems","status":"public","scopus_import":"1"}]