Ioannis Partalas

An Ensemble Pruning Primer

Ensemble pruning deals with the reduction of an ensemble of predictive models in order to improve its efficiency and predictive performance. The last 12 years a large number of ensemble pruning methods have been proposed. This work proposes a taxonomy for their organization and reviews important representative methods of each category. It abstracts their key components and discusses their main advantages and disadvantages. We hope that this work will serve as a good starting point and reference for researchers working on the development of new ensemble pruning methods.

An overview of the BIOASQ large-scale biomedical semantic indexing and question answering competition

BackgroundThis article provides an overview of the first BioASQ challenge, a competition on large-scale biomedical semantic indexing and question answering (QA), which took place between March and September 2013. BioASQ assesses the ability of systems to semantically index very large numbers of biomedical scientific articles, and to return concise and user-understandable answers to given natural language questions by combining information from biomedical articles and ontologies.ResultsThe 2013 BioASQ competition comprised two tasks, Task 1a and Task 1b. In Task 1a participants were asked to automatically annotate new PubMed documents with MeSH headings. Twelve teams participated in Task 1a, with a total of 46 system runs submitted, and one of the teams performing consistently better than the MTI indexer used by NLM to suggest MeSH headings to curators. Task 1b used benchmark datasets containing 29 development and 282 test English questions, along with gold standard (reference) answers, prepared by a team of biomedical experts from around Europe and participants had to automatically produce answers. Three teams participated in Task 1b, with 11 system runs. The BioASQ infrastructure, including benchmark datasets, evaluation mechanisms, and the results of the participants and baseline methods, is publicly available.ConclusionsA publicly available evaluation infrastructure for biomedical semantic indexing and QA has been developed, which includes benchmark datasets, and can be used to evaluate systems that: assign MeSH headings to published articles or to English questions; retrieve relevant RDF triples from ontologies, relevant articles and snippets from PubMed Central; produce “exact” and paragraph-sized “ideal” answers (summaries). The results of the systems that participated in the 2013 BioASQ competition are promising. In Task 1a one of the systems performed consistently better from the NLM’s MTI indexer. In Task 1b the systems received high scores in the manual evaluation of the “ideal” answers; hence, they produced high quality summaries as answers. Overall, BioASQ helped obtain a unified view of how techniques from text classification, semantic indexing, document and passage retrieval, question answering, and text summarization can be combined to allow biomedical experts to obtain concise, user-understandable answers to questions reflecting their real information needs.

Pruning an ensemble of classifiers via reinforcement learning

This paper studies the problem of pruning an ensemble of classifiers from a reinforcement learning perspective. It contributes a new pruning approach that uses the Q-learning algorithm in order to approximate an optimal policy of choosing whether to include or exclude each classifier from the ensemble. Extensive experimental comparisons of the proposed approach against state-of-the-art pruning and combination methods show very promising results. Additionally, we present an extension that allows the improvement of the solutions returned by the proposed approach over time, which is very useful in certain performance-critical domains.

An ensemble uncertainty aware measure for directed hill climbing ensemble pruning

This paper proposes a new measure for ensemble pruning via directed hill climbing, dubbed Uncertainty Weighted Accuracy (UWA), which takes into account the uncertainty of the decision of the current ensemble. Empirical results on 30 data sets show that using the proposed measure to prune a heterogeneous ensemble leads to significantly better accuracy results compared to state-of-the-art measures and other baseline methods, while keeping only a small fraction of the original models. Besides the evaluation measure, the paper also studies two other parameters of directed hill climbing ensemble pruning methods, the search direction and the evaluation dataset, with interesting conclusions on appropriate values.

Greedy regression ensemble selection: Theory and an application to water quality prediction

This paper studies the greedy ensemble selection family of algorithms for ensembles of regression models. These algorithms search for the globally best subset of regressors by making local greedy decisions for changing the current subset. We abstract the key points of the greedy ensemble selection algorithms and present a general framework, which is applied to an application domain with important social and commercial value: water quality prediction.

BioASQ: A Challenge on Large-Scale Biomedical Semantic Indexing and Question Answering

BioASQ is a series of challenges that aims to assess the performance of information systems in supporting two tasks that are central to the biomedical question answering process: a the indexing of large volumes of unlabelled data, primarily scientific articles, with biomedical concepts, b the processing of biomedical questions and the generation of answers and supporting material. In this paper, the main results of the first two BioASQ challenges are presented.

Evaluation measures for hierarchical classification: a unified view and novel approaches

Hierarchical classification addresses the problem of classifying items into a hierarchy of classes. An important issue in hierarchical classification is the evaluation of different classification algorithms, an issue which is complicated by the hierarchical relations among the classes. Several evaluation measures have been proposed for hierarchical classification using the hierarchy in different ways without however providing a unified view of the problem. This paper studies the problem of evaluation in hierarchical classification by analysing and abstracting the key components of the existing performance measures. It also proposes two alternative generic views of hierarchical evaluation and introduces two corresponding novel measures. The proposed measures, along with the state-of-the-art ones, are empirically tested on three large datasets from the domain of text classification. The empirical results illustrate the undesirable behaviour of existing approaches and how the proposed methods overcome most of these problems across a range of cases.

Ensemble pruning using reinforcement learning

Multiple Classifier systems have been developed in order to improve classification accuracy using methodologies for effective classifier combination. Classical approaches use heuristics, statistical tests, or a meta-learning level in order to find out the optimal combination function. We study this problem from a Reinforcement Learning perspective. In our modeling, an agent tries to learn the best policy for selecting classifiers by exploring a state space and considering a future cumulative reward from the environment. We evaluate our approach by comparing with state-of-the-art combination methods and obtain very promising results.

Transfer learning in multi-agent reinforcement learning domains

In the context of reinforcement learning, transfer learning refers to the concept of reusing knowledge acquired in past tasks to speed up the learning procedure in new tasks. Transfer learning methods have been succesfully applied in single-agent reinforcement learning algorithms, but no prior work has focused on applying them in a multi-agent environment. We propose a novel method for transfer learning in multi-agent reinforcement learning domains. We proceed to test the proposed approach in a multi-agent domain under various configurations. The results demonstrate that the method can reduce the learning time and increase the asymptotic performance of the learning algorithm.

Transferring task models in Reinforcement Learning agents

The main objective of transfer learning is to reuse knowledge acquired in a previous learned task, in order to enhance the learning procedure in a new and more complex task. Transfer learning comprises a suitable solution for speeding up the learning procedure in Reinforcement Learning tasks. This work proposes a novel method for transferring models to Reinforcement Learning agents. The models of the transition and reward functions of a source task, will be transferred to a relevant but different target task. The learning algorithm of the target tasks agent takes a hybrid approach, implementing both model-free and model-based learning, in order to fully exploit the presence of a source task model. Moreover, a novel method is proposed for transferring models of potential-based reward shaping functions. The empirical evaluation, of the proposed approaches, demonstrated significant results and performance improvements in the 3D Mountain Car and Server Job Scheduling tasks, by successfully using the models generated from their corresponding source tasks.