On the occasion of World Logic Day, January 14, 2025, the DLMPST Commission on Logic Education invites you to a webinar by Professor Moshe Vardi.
Model explainability has become an important problem in artificial intelligence (AI) due to the increased effect that algorithmic predictions have on humans. Explanations can help users understand not only why AI models make certain predictions, but also how these predictions can be changed via counterfactual explanations. Given a data point and a trained model, we want to find the minimal perturbation to the input such that the prediction changes. We frame the problem of finding counterfactual explanations as a gradient-based optimization task and first focus on tree ensembles. We then extend our method to accommodate graph neural networks (GNNs), given the increasing promise of GNNs in real-world applications such as fake news detection and molecular simulation.
Explainable artificial intelligence (XAI) is a mainstay of trustworthy AI. Recent years have witnessed massive efforts towards delivering some sort of XAI solutions. Most of these efforts are based on non-symbolic methods, and invariably will produce erroneous results. As a result, even if the predictions of a machine learning model could be trusted, the lack of reliable explanations will also make those predictions unworthy of trust. This talk provides a brief glimpse of the emerging field of logic-based explainable AI, a rigorous alternative to the still widely-used but extremely problematic non-symbolic methods.
Register for free at: https://tuwien.zoom.us/webinar/register/WN_msQDcDJbRfWQOUwdIDqzGQ
Marco Scarsini (Luiss University, Rome) will be visiting us from January 16-17, 2025 and give a talk at CWI on Thursday, January 16, 11:00-12:00, room L016.
This months episode of Talk That Science will feature special guest Erik Stei, Assistant Professor (Universitair docent) at Utrecht University working mainly in the philosophy of logic, in the philosophy of language, and in epistemology and the philosophy of science.
Researchers of any nationality with seven to twelve years of experience since the completion of their PhD, a scientific track record showing great promise, and an excellent research proposal can apply. Extensions are possible. Consolidator Grants may be awarded up to €2 million for a period of up to five years. The deadline is 14 January 2025. The call is open as of 26 September 2024.
You can now submit nominations for the KHMW Dissertation Prize Interdisciplinarity 2025, honoring research with a distinctive interdisciplinary character that transcends the boundaries between the classical alpha/beta/gamma scientific domains. The €10,000 prize is intended as an encouragement for researchers who completed their Ph.D. between September 1, 2023, and August 31, 2024.
The prize, established since 2023 by the Dr. Elizabeth Schram-Mulley (ESM) Foundation, is awarded by the Royal Holland Society of Sciences. The foundation honors the memory of and manages the legacy of Dr. Schram-Mulley, who, as a publisher and benefactor, held research and education in high regard.
You can now submit submit nominations for the KHMW Keetje Hodshon Dissertation Prize for the Humanities, intended for research in the field of Language and Communication Sciences in 2025. The €10,000 prize is intended as an incentive prize for researchers who received their PhD in the period 1/1/2021 – 31/12/2024.
Do you have an innovative idea to tackle a societal challenge? Sign up for the Amsterdam Science & Innovation Awards 2025 (AmSIA) and stand a chance at winning €10,000 to develop your idea that contributes to a better world. The competition is open to researchers, PhD candidates, staff members and students.
The Department of Computer Science at Lund University invites applications for a PhD position in theoretical computer science and/or combinatorial optimization.
The PhD student will be working in the Mathematical Insights into Algorithms for Optimization (MIAO) group headed by Jakob Nordström, which is active at both the University of Copenhagen and Lund University on either side of the Oresund Bridge. The MIAO research group has a unique profile in that we are doing cutting-edge research both on the mathematical foundations of efficient computation and on state-of-the-art practical algorithms for real-world problems. This leads to classical questions in computational complexity theory—though often with new, fascinating twists—but also involves work on devising clever algorithms that can exploit the power of such paradigms in practice.
This is a four-year full-time employed position, but PhD positions usually (though not necessarily) include 20% teaching, in which case they are prolonged for one more year. The starting date is negotiable, but should ideally be in August-September 2025 or earlier. All positions in the research group are fully funded, employed positions (including travel money) that come with an internationally competitive salary.
There is a one year postdoctoral position at the University of Salento (Lecce,Italy) on the project Ethical Design for AI (eDefAI), funded by the EU, as part of the PNRR Project FAIR Spoke 6.
The research will be focused on the formalization of norms for fair and unbiased reasoning with data in AI. A particular focus will be given to norms that hinge on both epistemic and ethical aspects. The researcher will use formalisms either from the area of formal argumentation, or deontic logic or other non-classical logics. Competence in one or more of these areas is expected.
Please note that the call is only available in italian, but no knowledge of italian is required for the job. Furthermore, flexible working arrangements will be possible.
The Danish Data Science Academy invites applications for postdoc and PhD fellowships for visionary and ambitious young data scientists who want to pursue their own research ideas in collaboration with a Danish research environment. The postdoc call is at https://ddsa.dk/postdocfellowshipcall2025/ and the PhD call at https://ddsa.dk/phdfellowshipcall2025/, and the application deadline is March 5, 2025.
Applications can be within any field of data science aligning with the DDSA research scope, including, but not limited to, algorithms research within data science and AI and applications of data science or computer science to other areas in natural, technical, or life sciences. The positions will be awarded to the most promising candidates according to their scientific qualifications, motivation, and engagement, as well as the quality, originality, relevance and potential impact of the proposed project. It is a requirement that the applicant has a well-defined research project proposal as well as an agreement with a principal supervisor at a Danish university.
Informal inquiries about opportunities in the Algorithms and Complexity Section at the University of Copenhagen are welcome and can be directed to Jakob Nordstrom or other faculty in the section. Please contact us as soon as possible if you are interested, so that there is enough time to talk about your research ideas and provide feedback on how to turn them into a successful DDSA proposal (and also do not forget the open PhD call at https://jakobnordstrom.se/openings/PhD-Lund-250117.html closing on January 17, 2025).