PhD in Explainable AI: Attribution Methods for Text and Audio

Faculty/Services: Faculty of Science

Educational level: Master

Function type: PhD position

Closing date: 21 July 2022

Vacancy number: 9631

We are seeking a PhD candidate for the project 'InDeep: Interpreting Deep Learning Models for Text and Sound Methods and Applications' (NWO NWA 1292.19.399). This ambitious project aims to develop, apply, and fine-tune techniques to make modern deep learning models for text, speech, and music more transparent.

Would you like a PhD position where you do fundamental research, with immediate applicability, in Explainable AI? Where you work on evaluating and extending attribution methods for ‘opening the blackbox’ of deep learning models for text and audio processing? If you are excited about doing this kind of research in an interdisciplinary environment with smart and friendly colleagues and a strong industrial collaboration, then you may want to join us.

The PhD candidate will have supervision team consisting of Dr Willem Zuidema (NLP, Interpretability, cognitive modelling) and at least one other senior member of the InDeep consortium.

The project is funded by an NWO Dutch Research Agenda grant to a consortium led by Dr. Willem Zuidema of the Institute for Logic, Language, and Computation (ILLC) at the University of Amsterdam (UvA). The consortium also includes Tilburg University, the University of Groningen, Radboud University, and Vrije Universiteit Amsterdam as academic partners and KPN, AIGent, TNO, Textkernel, Chordify, GlobalTextware, Deloitte, and Floodtags as industrial partners.

What are you going to do

The goal of PhD project is to develop and fine-tune data-driven interpretability methods for DL models of language, speech and music. We start with the subclass of data-driven methods known as ‘attribution methods’, including those based on computing gradients, like LRP, and those based on computing ‘Shapley-values’, like SHAP and Contextual Decomposition. Key insights guiding the work are that (1) simplifications/approximations are always required when the DL-model is nonlinear, (2) that no single method is best for all usages. Rather, the details of the application determine which of the necessary simplifications are justified.

The ultimate goal of the project is to provide a comparison of the usefulness of different methods in many different applications, and deeper insights about why certain methods work are best for certain applications. We also aim to extend the range of data-driven interpretability methods, by combining and existing methods and by moving such methods beyond attributing responsibility to individual words (syllables/notes/chords). In particular, we work towards identifying more complex structure in the data that in concord leads to a particular classification or decision. Our overview of attribution methods, and their do’s and don’ts, will also play a key role in our Industry Outreach & Education programme.

Tasks and responsibilities:

• Independently carrying out research, including writing and publishing three to four peer-reviewed articles.

• Submitting a PhD thesis within the period of appointment.

• Participating in the PhD programme of the ILLC.

• Participating in and contributing to the organisation of research activities and events at the ILLC, such as workshops and colloquia.

• Making a small contribution to the ILLC’s educational mission by working as a teaching assistant for courses in your area of expertise and by assisting with the supervision of student research projects.

• Regularly presenting research results at international workshops and conferences, and publishing them in conference proceedings and journals.

What do we require of you

We expect you to have:

• a completed master’s degree in artificial intelligence, computer science or Natural Language Processing, or comparable.

• a serious interest in natural language, speech and/or music research

• an excellent academic track record

• working knowledge of state-of-the art techniques in deep learning (with knowledge of explainable AI techniques desirable but not required)

• good programming skills and experience with Python or R and PyTorch or TensorFlow

• full professional proficiency in spoken and written English

• the ability to finish the PhD thesis in four years (i.e., good skills in planning, taking initiative, and academic writing)

You may apply if you have not yet completed your master's degree only if you provide a signed letter from your supervisor stating that you will graduate before 1 November 2022.

Please note that knowledge of the Dutch language is not required for this position, nor is it required for being able to live in Amsterdam. However, if you wish, as a PhD candidate at the ILLC you will have the opportunity to attend Dutch language classes.

If you already hold a doctorate/PhD or are working towards obtaining a similar degree elsewhere, you will not be admitted to a doctoral programme at the UvA.

Our offer

A temporary contract for 38 hours per week for the duration of 4 years (the initial contract will be for a period of 18 months and after satisfactory evaluation it will be extended for a total duration of 4 years). The preferred starting date is 1 November 2022. This should lead to a dissertation (PhD thesis). We will draft an educational plan that includes attendance of courses and (international) meetings. We also expect you to assist in teaching undergraduates and master students.

The gross monthly salary, based on 38 hours per week and dependent on relevant experience, ranges between € 2,443 to € 3,122 (scale P). This does not include 8% holiday allowance and 8,3% year-end allowance. The Collective Labour Agreement of Universities of the Netherlands is applicable.

About us

The University of Amsterdam (UvA) is the Netherlands’ largest university, offering the widest range of academic programmes. At the UvA, 30,000 students, 6,000 staff members and 3,000 PhD candidates study and work in a diverse range of fields, connected by a culture of curiosity.

The Institute for Logic, Language and Computation (ILLC) is a research institute at the UvA in which researchers from the Faculty of Science and the Faculty of Humanities. collaborate. Its central research area is the study of fundamental principles of encoding, transmission and comprehension of information. Research at ILLC is interdisciplinary, and aims at bringing together insights from various disciplines concerned with information and information processing, such as logic, mathematics, computer science, linguistics, natural language processing, cognitive science, artificial intelligence, music cognition, and philosophy.

Any questions

Do you have any questions or do you require additional information? Please contact:

• Dr. Jelle Zuidema

Job application

Do you recognise yourself in the job profile? Then we look forward to receiving your application. You may apply online by using the link below. Applications should include the following information, in one PDF file (not zipped):

• a curriculum vitae (max two pages, font size 12), including a link to your Master’s thesis, and the names, affiliations and email addresses of two referees; and

• a letter of motivation (max one page, font size 12), explaining why you are a good fit to this position.

Please use the mandatory CV field in the application form to upload one single PDF containing the items listed above. Only the file uploaded in the CV field will be considered by the search committee. Do not upload any other attachment.

The response period closes on 21 July 2022. Only complete applications submitted as one pdf file — received within the response period via the link below — will be considered.

The interviews will be held in July-August 2022.

The UvA is an equal-opportunity employer. We prioritise diversity and are committed to creating an inclusive environment for everyone. We value a spirit of enquiry and perseverance, provide the space to keep asking questions, and promote a culture of curiosity and creativity.

If you encounter Error GBB451, reach out to our HR Department directly. They will gladly help you continue your application.