{"id":12,"date":"2023-06-15T14:53:06","date_gmt":"2023-06-15T14:53:06","guid":{"rendered":"https:\/\/projects.illc.uva.nl\/indeep\/?page_id=12"},"modified":"2023-07-21T11:21:16","modified_gmt":"2023-07-21T11:21:16","slug":"indeep-interpreting-deep-learning-models-for-text-and-sound","status":"publish","type":"page","link":"https:\/\/projects.illc.uva.nl\/indeep\/","title":{"rendered":"InDeep Interpreting Deep Learning Models for Text and Sound"},"content":{"rendered":"\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n

In the InDeep project pioneering researchers in the domain of interpretability of deep learning models of text, language, speech and music are brought together. They collaborate with companies and non-for-profit institutions working with language, speech and music technology, to develop applications that help assess the usefulness of alternative interpretability techniques on a range of different tasks. In \u201cjustification\u201d tasks, we look at how interpretability techniques help give users meaningful feedback. Examples include legal and medical document text mining and audio search. In \u201caugmentation\u201d tasks we look at how these techniques facilitate the use of domain knowledge and models from outside deep learning to make the models perform even better. Examples include machine translation, music recommendation and writing feedback. In \u201cinteraction\u201d tasks we allow users to influence the functioning of their automated systems, by providing both interpretable information on how the system operates, and letting human produced output find its way into the internal states of the learning algorithm. Examples include adapting speech recognition to non-standard accents and dialects, interactive music generation, and machine assisted translation.<\/p>\n\n\n\n

Activities<\/em><\/p>\n\n\n\n