The Regression Tsetlin Machine: A Tsetlin Machine for Continuous Output Problems

被引:11
作者
Abeyrathna, K. Darshana [1 ]
Granmo, Ole-Christoffer [1 ]
Jiao, Lei [1 ]
Goodwin, Morten [1 ]
机构
[1] Univ Agder, Ctr Artificial Intelligence Res, Grimstad, Norway
来源
PROGRESS IN ARTIFICIAL INTELLIGENCE, PT II | 2019年 / 11805卷
关键词
Tsetlin Machine; Regression Tsetlin Machine; Tsetlin Automata; Regression; Pattern recognition; Propositional logic;
D O I
10.1007/978-3-030-30244-3_23
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The recently introduced Tsetlin Machine (TM) has provided competitive pattern classification accuracy in several benchmarks, composing patterns with easy-to-interpret conjunctive clauses in propositional logic. In this paper, we go beyond pattern classification by introducing a new type of TMs, namely, the Regression Tsetlin Machine (RTM). In all brevity, we modify the inner inference mechanism of the TM so that input patterns are transformed into a single continuous output, rather than to distinct categories. We achieve this by: (1) using the conjunctive clauses of the TM to capture arbitrarily complex patterns; (2) mapping these patterns to a continuous output through a novel voting and normalization mechanism; and (3) employing a feedback scheme that updates the TM clauses to minimize the regression error. The feedback scheme uses a new activation probability function that stabilizes the updating of clauses, while the overall system converges towards an accurate input-output mapping. The performance of the RTM is evaluated using six different artificial datasets with and without noise, in comparison with the Classic Tsetlin Machine (CTM) and the Multiclass Tsetlin Machine (MTM). Our empirical results indicate that the RTM obtains the best training and testing results for both noisy and noise-free datasets, with a smaller number of clauses. This, in turn, translates to higher regression accuracy, using significantly less computational resources.
引用
收藏
页码:268 / 280
页数:13
相关论文
共 13 条
[1]  
Abeyrathna K., 2019, ARXIV190504199
[2]   A Novel Tsetlin Automata Scheme to Forecast Dengue Outbreaks in the Philippines [J].
Abeyrathna, K. Darshana ;
Granmo, Ole-Christoffer ;
Goodwin, Morten .
2018 IEEE 30TH INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE (ICTAI), 2018, :680-685
[3]   Stochastic Learning for SAT-Encoded Graph Coloring Problems [J].
Bouhmala, Noureddine ;
Granmo, Ole-Christoffer .
INTERNATIONAL JOURNAL OF APPLIED METAHEURISTIC COMPUTING, 2010, 1 (03) :1-19
[4]  
Goodwin M., 2018, ARXIV180904547
[5]  
Granmo O.-C., 2018, arXiv
[6]   Solving Stochastic Nonlinear Resource Allocation Problems Using a Hierarchy of Twofold Resource Allocation Automata [J].
Granmo, Ole-Christoffer ;
Oommen, B. John .
IEEE TRANSACTIONS ON COMPUTERS, 2010, 59 (04) :545-560
[7]   A simple learning automata-based solution for intrusion detection in wireless sensor networks [J].
Misra, Sudip ;
Krishna, P. Venkata ;
Abraham, Kiran Isaac .
WIRELESS COMMUNICATIONS & MOBILE COMPUTING, 2011, 11 (03) :426-441
[8]  
Narendra K.S., 2012, Learning Automata: An Introduction
[9]   LEARNING AUTOMATA - SURVEY [J].
NARENDRA, KS ;
THATHACHAR, MA .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1974, SMC4 (04) :323-334
[10]   A solution to the stochastic point location problem in metalevel nonstationary environments [J].
Oommen, B. John ;
Kim, Sang-Woon ;
Samuel, Mathew T. ;
Granmo, Ole-Christoffer .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2008, 38 (02) :466-476