|Appears in Collections:||Computing Science and Mathematics eTheses|
|Title:||Improving the Generalisability of Brain Computer Interface Applications via Machine Learning and Search-Based Heuristics|
Brain Computer Interface
Iterated Local Search
Brain Machine Interface
|Publisher:||University of Stirling|
|Citation:||Jason Adair, Alexander Brownlee, and Gabriela Ochoa. Evolutionary Algorithms with Linkage Information for Feature Selection in Brain Computer Interfaces. In Advances in Intelligent Systems and Computing, volume 513, pages 287–307. 2017. ISBN 9783319465616. doi: 10.1007/ 978-3-319-46562-3_19.|
Jason Adair, Alexander Brownlee, Fabio Daolio, and Gabriela Ochoa. Evolving training sets for improved transfer learning in brain computer interfaces. In Machine Learning, Optimization, and Big Data, pages 186–197. Springer International Publishing, 2018. ISBN 978-3-319-72926-8. doi: 10.1007/978-3-319-72926-8_16.
Jason Adair, Alexander E. I. Brownlee, and Gabriela Ochoa. Mutual information iterated local search: A wrapper-filter hybrid for feature selection in brain computer interfaces. In Applications of Evolutionary Computation, pages 63–77, Cham, 2018. Springer International Publishing. ISBN 978-3-319-77538-8. doi: 10.1007/978-3-319-77538-8_5.
|Abstract:||Brain Computer Interfaces (BCI) are a domain of hardware/software in which a user can interact with a machine without the need for motor activity, communicating instead via signals generated by the nervous system. These interfaces provide life-altering benefits to users, and refinement will both allow their application to a much wider variety of disabilities, and increase their practicality. The primary method of acquiring these signals is Electroencephalography (EEG). This technique is susceptible to a variety of different sources of noise, which compounds the inherent problems in BCI training data: large dimensionality, low numbers of samples, and non-stationarity between users and recording sessions. Feature Selection and Transfer Learning have been used to overcome these problems, but they fail to account for several characteristics of BCI. This thesis extends both of these approaches by the use of Search-based algorithms. Feature Selection techniques, known as Wrappers use ‘black box’ evaluation of feature subsets, leading to higher classification accuracies than ranking methods known as Filters. However, Wrappers are more computationally expensive, and are prone to over-fitting to training data. In this thesis, we applied Iterated Local Search (ILS) to the BCI field for the first time in literature, and demonstrated competitive results with state-of-the-art methods such as Least Absolute Shrinkage and Selection Operator and Genetic Algorithms. We then developed ILS variants with guided perturbation operators. Linkage was used to develop a multivariate metric, Intrasolution Linkage. This takes into account pair-wise dependencies of features with the label, in the context of the solution. Intrasolution Linkage was then integrated into two ILS variants. The Intrasolution Linkage Score was discovered to have a stronger correlation with the solutions predictive accuracy on unseen data than Cross Validation Error (CVE) on the training set, the typical approach to feature subset evaluation. Mutual Information was used to create Minimum Redundancy Maximum Relevance Iterated Local Search (MRMR-ILS). In this algorithm, the perturbation operator was guided using an existing Mutual Information measure, and compared with current Filter and Wrapper methods. It was found to achieve generally lower CVE rates and higher predictive accuracy on unseen data than existing algorithms. It was also noted that solutions found by the MRMR-ILS provided CVE rates that had a stronger correlation with the accuracy on unseen data than solutions found by other algorithms. We suggest that this may be due to the guided perturbation leading to solutions that are richer in Mutual Information. Feature Selection reduces computational demands and can increase the accuracy of our desired models, as evidenced in this thesis. However, limited quantities of training samples restricts these models, and greatly reduces their generalisability. For this reason, utilisation of data from a wide range of users is an ideal solution. Due to the differences in neural structures between users, creating adequate models is difficult. We adopted an existing state-of-the-art ensemble technique Ensemble Learning Generic Information (ELGI), and developed an initial optimisation phase. This involved using search to transplant instances between user subsets to increase the generalisability of each subset, before combination in the ELGI. We termed this Evolved Ensemble Learning Generic Information (eELGI). The eELGI achieved higher accuracy than user-specific BCI models, across all eight users. Optimisation of the training dataset allowed smaller training sets to be used, offered protection against neural drift, and created models that performed similarly across participants, regardless of neural impairment. Through the introduction and hybridisation of search based algorithms to several problems in BCI we have been able to show improvements in modelling accuracy and efficiency. Ultimately, this represents a step towards more practical BCI systems that will provide life altering benefits to users.|
|Type:||Thesis or Dissertation|
|thesis.pdf||PhD Thesis - Jason Adair March 2018||10.36 MB||Adobe PDF||View/Open|
This item is protected by original copyright
Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.
If you believe that any material held in STORRE infringes copyright, please contact firstname.lastname@example.org providing details and we will remove the Work from public display in STORRE and investigate your claim.