PhD Position in Geometric Deep Learning of Space and Time

PhD Position in Geometric Deep Learning of Space and Time
PhD Position in Geometric Deep Learning of Space and Time

Published	Deadline	Location
today	15 Nov	Amsterdam

Job description
Geometry Deep Learning has been one of the most impactful new paradigms in research in Deep Learning, Machine Learning, and Computer Vision. The study of manifolds, graphs, their geometric properties, their relation to data manifold hypothesis, equivariance, have advanced not only the theory of neural networks, but also applications. Specifically, these models have been increasingly used to model not only point clouds, 3D shapes, scene renderings, but also molecular and protein 3D surfaces. What is more, recently there have been quite interesting advances in the intersection of Deep Learning, Interacting Dynamical Systems, PDEs, and simulations, where geometry plays a pivotal role on defining the domain in which neural network functions are applied to (adaptive/resolution-free grids, n-dimensional settings), or how to best leverage the geometric properties of co-existing physical components.
In this position we will study the theory and applications of geometry in various aspects of deep neural networks, focusing on spatial, as well as on temporal data and dynamics. The idea is that geometry is also very important for physics-informed models, where the latent space adopts a geometrical form that conforms to a certain prior that we know must be relevant, towards what one could ‘physics-informed representation learning’ (learn representations that internally encode part of physical reality), as opposed to physics-informed neural networks (that try to imitate physical processes) and object-centric representation learning (which is devoid of any physical notion at the representational level). For instance, if you had the video of a single pendulum, can you learn a representation that not only encodes what the pendulum is, but also what the pendulum does, eg, by attempting to incorporate conservation laws as some form of geometric inductive priors, which the NN can pick up and learn its own version?
Some relevant research questions for this position will be:

• Can we re-interpret existing neural network models in the form of geometric primitives and components?
• Can we meaningfully use advanced geometric structures, in the form of graphs or manifolds, as components and latent spaces of deep neural networks?
• Can we use temporal dynamics to recover intricate and complex geometric properties in data and physical environments?
• Inspired by successful diffusion models, can we interpret random walks on complex geometries to define forward and backward propagation in space and time?
• Can we use advanced geometry to understand spatiotemporal and dynamical processes, like complex PDEs?
• Can geometry help with designing more parameter-efficient, intuitive, and explainable deep learning models?
• What are the best applications in sciences with geometry and deep learning?

The student will be supervised by Dr. Efstratios Gavves, Associate Professor at the University of Amsterdam, and Dr. Taco Cohen from Qualcomm. The position will be part of the QUVA Lab, which is an academic-industry lab supported by Qualcomm, and the ELLIS Network of Excellence in AI.
What are you going to do?
You will carry out research and development in the area of Deep Machine Learning and Vision. The research is embedded in the QUVA Lab and VISlab group at the University of Amsterdam.
Your tasks will be to:

• Develop new deep machine learning and/or computer vision methods on Geometric Deep Learning of Space and Time;
• Collaborate with other researchers within the lab;
• Regularly present internally on your progress;
• Regularly present intermediate research at international conferences and workshops, publish them in proceedings and journals, help with submitting applications to protect IP;
• Assist in relevant teaching activities;
• Complete and defend a PhD thesis within the official appointment duration of four years.

Specifications

• max. 38 hours per week
• €2541—€3247 per month
• Amsterdam View on Google Maps

University of Amsterdam (UvA)

Requirements
Your experience and profile:

• An MSc degree in Artificial Intelligence, (Applied) Mathematics/Physics, Computer Science, Engineering or related field;
• A strong background/knowledge in machine learning and statistics. A background/ knowledge in computer vision is also a plus;
• A strong background/knowledge in geometry, and stochastic differential equations;
• Excellent programming skills preferably in Python;
• Solid mathematics foundations, especially statistics, calculus and linear algebra;
• You are highly motivated, independent, and creative;
• Strong communication, presentation and writing skills and excellent command of English.

Prior publications in relevant machine learning, or computer vision conferences or journals is advantageous.
Conditions of employment
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 as soon as possible. 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 € 2,541 to € 3,247 (scale P). This does not include the 8% holiday allowance and the 8,3% year-end allowance the UvA offers. The UFO profile PhD Candidate is applicable. A favourable tax agreement, the ‘30% ruling’, may apply to non-Dutch applicants. The Collective Labour Agreement of Universities of the Netherlands is applicable.
Besides the salary and a vibrant and challenging environment at Science Park we offer you multiple fringe benefits:

• 232 holiday hours per year (based on fulltime) and extra holidays between Christmas and 1 January.
• Multiple courses to follow from our Teaching and Learning Centre.
• A complete educational program for PhD students.
• Multiple courses on topics such as leadership for academic staff.
• Multiple courses on topics such as time management, handling stress and an online learning platform with 100+ different courses.
• 7 weeks birth leave (partner leave) with 100% salary.
• Partly paid parental leave.
• The possibility to set up a workplace at home;
• A pension at ABP for which UvA pays two third part of the contribution.
• The possibility to follow courses to learn Dutch;
• Help with housing for a studio or small apartment when you’re moving from abroad.

Are you curious to read more about our extensive package of secondary employment benefits, take a look here .
Employer
Faculty of Science
The University of Amsterdam 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 Faculty of Science has a student body of around 8,000, as well as 1,800 members of staff working in education, research or support services. Researchers and students at the Faculty of Science are fascinated by every aspect of how the world works, be it elementary particles, the birth of the universe or the functioning of the brain.
The mission of the Informatics Institute (IvI) is to perform curiosity-driven and use-inspired fundamental research in Computer Science. The main research themes are Artificial Intelligence, Computational Science and Systems and Network Engineering. Our research involves complex information systems at large, with a focus on collaborative, data driven, computational and intelligent systems, all with a strong interactive component.
The position is in the QUVA Lab with Dr. Efstratios Gavves, Associate Professor in the Video & Image Sense lab (VIS) led by Prof. C. Snoek, and Dr. Taco Cohen at Qualcomm . VIS is a world-leading lab on Computer Vision and Machine Learning, and has over 40 PhD students, postdoctoral researchers and faculty members working on a broad variety of deep learning, computer vision, and machine learning subjects: from stochastic probabilistic models, temporal causality, graph neural networks, neural networks dynamical systems, to action and object recognition, to efficient spatiotemporal deep learning, motion forecasting. Also, the position is also embedded in the ELLIS Network of Excellence in AI.
The QUVA Lab is embedded in the Video & Image Sense lab (VIS) and the Amsterdam Machine Learning lab (AMlab), two groups within the Informatics Institute working on advanced artificial intelligence. QUVA Lab houses several projects, from Federated Learning, Deep Compression, Combinatorial Optimization, Causal Representations Learning, to Video Action Recognition. The research will be done in collaboration with experts from Qualcomm AI Research. The lab is part of the Innovation Center for Artificial Intelligence , a Netherlands initiative focused on joint technology development between academia, industry and government in the area of artificial intelligence.
In the lab we encourage strongly collaborations. Other labs on Machine Learning and Computer Vision at the Informatics Institute include AMLab by Prof. M. Welling and Assoc. Prof. J. W. van de Meent, and Computer Vision lab (CV) by Prof. Th. Gevers.
Want to know more about our organisation? Read more about working at the University of Amsterdam.
Any questions?
Do you have any questions or do you require additional information? Please contact:

• Prof. Efstratios Gavves , Associate Professor

Working at UvA

The University of Amsterdam is ambitious, creative and committed: a leader in international science and a partner in innovation, the UvA has been inspiring generations since 1632.

Learn more

Apply for this job
Apply for this job

This application process is managed by the employer (University of Amsterdam (UvA)). Please contact the employer for questions regarding your application.

Apply for this job via the employer’s website
Thank you for applying

Please contact the employer for questions regarding your application.

Tip: save this job as favorite in your AcademicTransfer account. This gives you an immediate overview and makes it easy to find the job later on. No account yet? Create it now and take advantage of other useful functionalities too!

Back to the vacancy
Application procedure
If you feel the profile fits you, and you are interested in the job, we look forward to receiving your application. You can apply online via the button below. We accept applications until and including 15 November 2022.
Applications should include the following information (all files besides your CV should be submitted in one single pdf file):

• A motivation letter that motivates your choice for this position, and in particular, for which of the research project(s) you are applying;
• A curriculum vitae, including your list of publications if applicable;
• A research statement on how to approach the project of your choice. Solid and creative ideas will be greatly appreciated. (max 2 pages).
• A link to your Master’s thesis;
• A complete record of Bachelor and Master courses (including grades and explanation of grading system);
• A list of projects and publications you have worked on, with brief descriptions of your contributions (max 2 pages);
• The names and contact addresses of at least two academic references (please do not include any recommendation letters).

Please make sure to provide ALL requested documents mentioned above.
You can use the CV field to upload your resume as a separate pdf document. Use the Cover Letter field to upload the other requested documents, including the motivation letter, as one single pdf file.
Only complete applications received within the response period via the link below will be considered.
Candidate selection is done in cooperation with the ELLIS Network of Excellence in AI. The interviews will be held in the course of January 2023.