Visualising state space representations of LSTM networks

Smith, Emmanuel M.; Smith, Jim; Legg, Phil; Francis, Simon

Abstract

Long Short-Term Memory (LSTM) networks have proven to be one of the most effective models for making predictions on sequence-based tasks. These models work by capturing, remembering, and forgetting information relevant to their future predictions. The non-linear complexity of the mechanisms involved in this process means we currently lack tools for achieving interpretability. Ideally, we want these models to provide an explanation of why they make a particular prediction, given a specific input. Researchers have explored the idea of interpreting LSTMs in specific contexts such as natural language processing or classification, but they put minimal focus on approaches which are generalisable across different applications. To alleviate this, in this work, we demonstrate a method which enables the interpretation and comparison of LSTM states during time series predictions. We show that by reducing the dimensionality of network states one can scalably visualise patterns and explain model behaviours.

• Presentation Conference Type: Other
• Conference Name: Workshop on Visualization for AI Explainability
• Acceptance Date: Aug 2, 2018
• Deposit Date: Sep 19, 2018
• Publicly Available Date: Sep 19, 2018
• Peer Reviewed: Not Peer Reviewed
• Public URL: https://uwe-repository.worktribe.com/output/863245
• Additional Information: Title of Conference or Conference Proceedings : Workshop on Visualization for AI Explainability
• Contract Date: Sep 19, 2018