Learning alternative ways of performing a task

• Published in: Expert systems with applications 2020-06, Vol.148, p.113263, Article 113263
• Main Authors: Nieves, D., Ramírez-Quintana, MJ, Monserrat, C., Ferri, C., Hernández-Orallo, J.
• Format: Article
• Language: English
• Subjects: Activity recognition; Cooking; Identifying strategies; Inductive learning; Iterative methods; Machine learning; Process mining; Task learning; Training
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2020.113263

•Learning from experts requires learning from small and only positive examples.•We present an inductive approach for learning multiple models of tasks.•It can learn from few and only positive examples.•Traditional machine learning techniques are not useful in such scenarios.•Small set of meaningful examples is enough to obtain these models. A common way of learning to perform a task is to observe how it is carried out by experts. However, it is well known that for most tasks there is no unique way to perform them. This is especially noticeable the more complex the task is because factors such as the skill or the know-how of the expert may well affect the way she solves the task. In addition, learning from experts also suffers of having a small set of training examples generally coming from several experts (since experts are usually a limited and expensive resource), being all of them positive examples (i.e. examples that represent successful executions of the task). Traditional machine learning techniques are not useful in such scenarios, as they require extensive training data. Starting from very few executions of the task presented as activity sequences, we introduce a novel inductive approach for learning multiple models, with each one representing an alternative strategy of performing a task. By an iterative process based on generalisation and specialisation, we learn the underlying patterns that capture the different styles of performing a task exhibited by the examples. We illustrate our approach on two common activity recognition tasks: a surgical skills training task and a cooking domain. We evaluate the inferred models with respect to two metrics that measure how well the models represent the examples and capture the different forms of executing a task showed by the examples. We compare our results with the traditional process mining approach and show that a small set of meaningful examples is enough to obtain patterns that capture the different strategies that are followed to solve the tasks.