| layout |
|---|
default |
Deploying AI and ML systems to assist decision-making in real-world and safety-critical scenarios, requires different forms of complex reasoning under uncertainty. These scenarios include applications in healthcare and finance, as well as when certifying the fairness, robustness and privacy of ML systems. In all these cases, reasoning needs to be reliable and efficient and flexible enough to deal with constraints and background knowledge. The now consolidated field of tractable probabilistic models (TPMs) offers a very appealing approach as TPMs allow for exact inference or come with approximation guarantees, thus providing reliability while still allowing for efficient reasoning for a wide range of tasks, by design.
Furthermore, many TPMs provide a natural way to represent logical constraints and principled ways to incorporate them into larger ML systems. The spectrum of TPMs consists of a wide variety of techniques including models with tractable likelihoods (e.g., normalizing flow and autoregressive models), tractable marginals (e.g., bounded-treewidth models and determinantal point processes), and more complex tractable reasoning tasks (e.g., probabilistic and logic circuits, tensor networks).
This new edition of the Tractable Probabilistic Modeling workshop focuses on scaling and providing guarantees when used for probabilistic and logical reasoning, e.g., in neuro-symbolic AI where agents have to model both calibrated uncertainties and satisfy given background knowledge while being efficient. We also welcome contributions around the TPM spectrum that consider only one of the two aspects, and hope to bring the two communities together to advance their respective fields.
We invite three types of submissions:
- Novel research on tractable probabilistic modeling
- Retrospective papers discussing the impact, consequences, and lessons learned from classical TPM papers
- Recently accepted papers at top ML / AI conferences (in the original format and length)
Topics of interest include, but are not limited to:
- New tractable representations in logical, continuous, and hybrid domains
- Learning algorithms for TPMs
- Theoretical and empirical analysis of TPMs
- Connections between TPM classes
- TPMs for responsible, robust, and explainable AI
- Approximate inference algorithms with guarantees
- Successful applications of TPMs to real-world problems,with a special focus on NeSy AI
Original papers and retrospective papers are required to follow the style guidelines of UAI. Submitted papers should be up to 4 pages long, excluding references. Already accepted papers can be submitted in the format of the venue they have been accepted to. Supplementary material can be put in the same pdf paper (after references); it is entirely up to the reviewers to decide whether they wish to consult this additional material.
All submissions must be electronic (through the link below), and must closely follow the formatting guidelines in the templates, otherwise, they will automatically be rejected. Reviewing for TPM is single-blind; i.e., reviewers will know the authors’ identity but authors won't know the reviewers' identity. However, we recommend that you refer to your prior work in the third person wherever possible. We also encourage links to public repositories such as GitHub to share code and/or data.
For any questions, please contact us at: [email protected]
Submission using: OpenReview
Note: New OpenReview profiles created without an institutional email will go through a moderation process that can take up to two weeks. OpenReview profiles created with an institutional email will be activated automatically.
All times are end of day AOE.
- {{ item.item }}: {{ item.date }}
{% for item in site.deadlines %}
{% endfor %}
- 7th Workshop on Tractable Probabilistic Modeling (2024)
- 6th Workshop on Tractable Probabilistic Modeling (2023)
- 5th Workshop on Tractable Probabilistic Modeling (2022)
- 4th Workshop on Tractable Probabilistic Modeling (2021)
- 3th Workshop on Tractable Probabilistic Modeling (2019)
- 2th Workshop on Tractable Probabilistic Modeling (2018)
- {{ item.name }}, {{ item.affiliation }} {% else %}
- {{ item.name }}, {{ item.affiliation }} {% endif %}
{% for item in site.data.organizers %}
{% if item.url %}
{% endfor %}