Filters








412 Hits in 6.1 sec

Optimal attribute-efficient learning of disjunction, parity, and threshold functions [chapter]

Ryuhei Uehara, Kensei Tsuchida, Ingo Wegener
1997 Lecture Notes in Computer Science  
More precise and often optimal results are presented for the cases where g is one of the functions disjunction, parity or threshold. 0  ...  k inputs by given constants this problem is known as attribute-e cient learning with k essential attributes. Results on general classes of functions are known.  ...  The functions disjunction, parity, and threshold are the most important basic gates in circuit theory and, therefore, the most fundamental functions.  ... 
doi:10.1007/3-540-62685-9_15 fatcat:ybfnfrbiyrbrpgbavczc53nsdu

Page 7405 of Mathematical Reviews Vol. , Issue 2001J [page]

2001 Mathematical Reviews  
We study the query complexity of attribute-efficient learning for three function classes that are, respectively, obtained from disjunction, parity, and threshold functions.  ...  ) Identification of partial disjunction, parity, and threshold functions.  ... 

Toward Attribute Efficient Learning of Decision Lists and Parities [chapter]

Adam R. Klivans, Rocco A. Servedio
2004 Lecture Notes in Computer Science  
We consider two well-studied problems regarding attribute efficient learning: learning decision lists and learning parity functions.  ...  For a wide range of parameters our construction matches a lower bound due to Beigel for decision lists and gives an essentially optimal tradeoff between polynomial threshold function degree and weight.  ...  Acknowledgements We thank Les Valiant for his observation that Claim 4.1 can be reinterpreted in terms of polynomial threshold functions, and we thank Jean Kwon for suggesting the Chebychev polynomial.  ... 
doi:10.1007/978-3-540-27819-1_16 fatcat:oo5c74vxtna3hakgzwn644wvoa

Toward Attribute Efficient Learning Algorithms [article]

Adam R. Klivans, Rocco A. Servedio
2003 arXiv   pre-print
Our approach establishes a relationship between attribute efficient learning and polynomial threshold functions and is based on a new construction of low degree, low weight polynomial threshold functions  ...  For a wide range of parameters our construction matches a 1994 lower bound due to Beigel for the ODDMAXBIT predicate and gives an essentially optimal tradeoff between polynomial threshold function degree  ...  Acknowledgements We thank Les Valiant for his observation that Claim 5 can be reinterpreted in terms of polynomial threshold functions. We thank Jean Kwon for suggesting the Chebychev polynomial.  ... 
arXiv:cs/0311042v1 fatcat:ra2td5h4gvdcxitxq4kiqo5ppa

Page 4584 of Mathematical Reviews Vol. , Issue 98G [page]

1998 Mathematical Reviews  
of disjunction, parity, and threshold functions.  ...  Results on general classes of functions are known. More precise and often optimal results are presented for the cases where g is one of the functions disjunction, parity or threshold.”  ... 

Exchangeable Variable Models [article]

Mathias Niepert, Pedro Domingos
2014 arXiv   pre-print
We prove that a family of tractable EVMs is optimal under zero-one loss for a large class of functions, including parity and threshold functions, and strictly subsumes existing tractable independence-based  ...  Extensive experiments show that EVMs outperform state of the art classifiers such as SVMs and probabilistic models which are solely based on independence assumptions.  ...  The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of ARO, ONR,  ... 
arXiv:1405.0501v1 fatcat:omnsozbrjff2djvie7drxh7f4a

Revising threshold functions

Robert H. Sloan, Balázs Szörényi, György Turán
2007 Theoretical Computer Science  
We give an efficient revision algorithm in the model of learning with equivalence and membership queries for threshold functions, and some negative results showing, for instance, that threshold functions  ...  cannot be revised efficiently from either type of query alone.  ...  The first and third author's researches were supported in part by the National Science Foundation under grants CCR-0100336 and CCF-0431059.  ... 
doi:10.1016/j.tcs.2007.03.034 fatcat:chatxvzgzvbwlkump6y6rgpg5i

Evolvability need not imply learnability [article]

Nisheeth Srivastava
2009 arXiv   pre-print
The implications of the latter case on the prospects of learning in complex hypothesis spaces is briefly examined.  ...  We show that Boolean functions expressible as monotone disjunctive normal forms are PAC-evolvable under a uniform distribution on the Boolean cube if the hypothesis size is allowed to remain fixed.  ...  Represent the space of attributes X = {0, 1} n , a set C of functions of the attributes, a set R of representations of functions 1 and a distribution D over the set X .  ... 
arXiv:0904.0648v1 fatcat:qbknbnzlsjhpfjz7gh47wmma5q

New Revision Algorithms [chapter]

Judy Goldsmith, Robert H. Sloan, Balázs Szörényi, György Turán
2004 Lecture Notes in Computer Science  
We also present an efficient revision algorithm for threshold functions.  ...  Attribute-efficient learning algorithms are required to be efficient (polynomial) in the number of relevant variables, and "super-efficient" (polylogarithmic) in the total number of variables [1, 2] .  ...  Valiant's related models [17, 18] also involve threshold functions, and as threshold functions are also known to be attribute-efficiently learnable, this raises the question whether threshold functions  ... 
doi:10.1007/978-3-540-30215-5_30 fatcat:iktnqh37lrd7hkdxpvp6w2bili

Rough fuzzy MLP: knowledge encoding and classification

M. Banerjee, S. Mitra, S.K. Pal
1998 IEEE Transactions on Neural Networks  
Results on classification of speech and synthetic data demonstrate the superiority of the system over the fuzzy and conventional versions of the MLP (involving no initial knowledge).  ...  The syntax of these rules automatically determines the appropriate number of hidden nodes while the dependency factors are used in the initial weight encoding.  ...  Note that the main motive of introducing this threshold function lies in reducing the size of the resulting network.  ... 
doi:10.1109/72.728363 pmid:18255803 fatcat:spj45i75ojgsvfced3lewmbedu

Tractable Learning for Complex Probability Queries

Jessa Bekker, Jesse Davis, Arthur Choi, Adnan Darwiche, Guy Van den Broeck
2015 Neural Information Processing Systems  
This allows us to support a broader class of complex probability queries, including counting, threshold, and parity, in polytime.  ...  Tractable learning aims to learn probabilistic models where inference is guaranteed to be efficient.  ...  AC and AD are partially supported by NSF (#IIS-1514253) and ONR (#N00014-12-1-0423).  ... 
dblp:conf/nips/BekkerDCDB15 fatcat:rxlcxzngmbf5ngcszxdml6ede4

Justicia: A Stochastic SAT Approach to Formally Verify Fairness

Bishwamittra Ghosh, Debabrota Basu, Kuldeep S. Meel
2021 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
As a technology ML is oblivious to societal good or bad, and thus, the field of fair machine learning has stepped up to propose multiple mathematical definitions, algorithms, and systems to ensure different  ...  Justicia is scalable, accurate, and operates on non-Boolean and compound sensitive attributes unlike existing distribution-based verifiers, such as FairSquare and VeriFair.  ...  Acknowledgments We are grateful to Jie-Hong Roland Jiang and Teodora Baluta for the useful discussion at the earlier stage of this project.  ... 
doi:10.1609/aaai.v35i9.16925 fatcat:5fqpoirfzjg47jnr64kntswo2u

Page 3847 of Mathematical Reviews Vol. , Issue 98F [page]

1998 Mathematical Reviews  
monotone k-term DNF (162-170); Ryuhei Uehara, Kensei Tsuchida and Ingo Wegener, Optimal attribute-efficient learning of disjunction, parity, and threshold functions (171-184); Satoshi Matsumoto and Ayumi  ...  with high prob- ability (66-78); Juris Smotrovs, Closedness properties in team learning of recursive functions (79-93); Matthias Ott and Frank Stephan, Structural measures for games and process control  ... 

Learning Coverage Functions and Private Release of Marginals [article]

Vitaly Feldman, Pravesh Kothari
2014 arXiv   pre-print
We study the problem of approximating and learning coverage functions.  ...  Alternatively, coverage functions can be described as non-negative linear combinations of monotone disjunctions.  ...  Acknowledgements We are grateful to Jan Vondrak for helpful advice and numerous discussions about this work.  ... 
arXiv:1304.2079v3 fatcat:lor5pjhisfd4vo534nxcwml2im

Justicia: A Stochastic SAT Approach to Formally Verify Fairness [article]

Bishwamittra Ghosh, Debabrota Basu, Kuldeep S. Meel
2021 arXiv   pre-print
As a technology ML is oblivious to societal good or bad, and thus, the field of fair machine learning has stepped up to propose multiple mathematical definitions, algorithms, and systems to ensure different  ...  Justicia is scalable, accurate, and operates on non-Boolean and compound sensitive attributes unlike existing distribution-based verifiers, such as FairSquare and VeriFair.  ...  Debabrota Basu was funded by WASP-NTU grant of the Knut and Alice Wallenberg Foundation during the initial phase of this work.  ... 
arXiv:2009.06516v2 fatcat:reyl3xmhyzbrhcia5eofllnxci
« Previous Showing results 1 — 15 out of 412 results