Similarity-Based Alignment and Generalization: A New Paradigm for Programming by Demonstration
نویسندگان
چکیده
We present an approach to learning procedural knowledge by demonstration called similarity-based alignment and generalization. Key to our approach is the ability to induce complex procedure structure (loops and conditional branches) by aligning multiple unannotated demonstrations of a procedure. We present an implemented instance of a similarity-based alignment and generalization algorithm that relies on the known Input-Output Hidden Markov Models, and describe an extension, the SimIOHMM, that significantly improves the algorithm’s performance. We present an empirical evaluation that demonstrates our system’s scaling performance and quantifies the performance increase obtained through the use of the SimIOHMM extension.
منابع مشابه
A generalization of Profile Hidden Markov Model (PHMM) using one-by-one dependency between sequences
The Profile Hidden Markov Model (PHMM) can be poor at capturing dependency between observations because of the statistical assumptions it makes. To overcome this limitation, the dependency between residues in a multiple sequence alignment (MSA) which is the representative of a PHMM can be combined with the PHMM. Based on the fact that sequences appearing in the final MSA are written based on th...
متن کاملSimilarity-Based Alignment and Generalization
We present a novel approach to learning predictive sequential models, called similarity-based alignment and generalization, which incorporates in the induction process a specific form of domain knowledge derived from a similarity metric of the points in the input space. When applied to Hidden Markov Models, our framework yields a new class of learning algorithms called SimAlignGen. We discuss t...
متن کاملgpALIGNER: A Fast Algorithm for Global Pairwise Alignment of DNA Sequences
Bioinformatics, through the sequencing of the full genomes for many species, is increasingly relying on efficient global alignment tools exhibiting both high sensitivity and specificity. Many computational algorithms have been applied for solving the sequence alignment problem. Dynamic programming, statistical methods, approximation and heuristic algorithms are the most common methods appli...
متن کاملAn Application of the ABS LX Algorithm to Multiple Sequence Alignment
We present an application of ABS algorithms for multiple sequence alignment (MSA). The Markov decision process (MDP) based model leads to a linear programming problem (LPP), whose solution is linked to a suggested alignment. The important features of our work include the facility of alignment of multiple sequences simultaneously and no limit for the length of the sequences. Our goal here is to ...
متن کاملGoal Programming Optimization Model for Performance Management: A SCOR-Based Supply Chain Decision Alignment
This article develops an integrated model of transmitting strategies and operational activities to enhance the efficiency of supply chain management. As the second objective, this paper aims to improve supply chain performance management (SCPM) by employing proper decision-making approaches. The proposed model optimizes the performance indicator based on SCOR metrics. A process-based method is ...
متن کامل