![[2602.14890] Lifted Relational Probabilistic Inference via Implicit Learning](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.14890] Lifted Relational Probabilistic Inference via Implicit Learning
Summary
This paper presents a novel approach to lifted relational probabilistic inference, integrating inductive learning and deductive reasoning without requiring explicit model construction.
Why It Matters
The research addresses a critical challenge in AI by merging learning and reasoning in first-order relational domains. This advancement could enhance the efficiency of probabilistic reasoning in AI applications, particularly in scenarios with incomplete or noisy data.
Key Takeaways
- Introduces a polynomial-time framework for implicit learning in first-order probabilistic logic.
- Combines incomplete axioms with partially observed examples for enhanced inference.
- Implements two simultaneous lifts: grounding-lift and world-lift for improved efficiency.
Computer Science > Artificial Intelligence arXiv:2602.14890 (cs) [Submitted on 16 Feb 2026] Title:Lifted Relational Probabilistic Inference via Implicit Learning Authors:Luise Ge, Brendan Juba, Kris Nilsson, Alison Shao View a PDF of the paper titled Lifted Relational Probabilistic Inference via Implicit Learning, by Luise Ge and 3 other authors View PDF HTML (experimental) Abstract:Reconciling the tension between inductive learning and deductive reasoning in first-order relational domains is a longstanding challenge in AI. We study the problem of answering queries in a first-order relational probabilistic logic through a joint effort of learning and reasoning, without ever constructing an explicit model. Traditional lifted inference assumes access to a complete model and exploits symmetry to evaluate probabilistic queries; however, learning such models from partial, noisy observations is intractable in general. We reconcile these two challenges through implicit learning to reason and first-order relational probabilistic inference techniques. More specifically, we merge incomplete first-order axioms with independently sampled, partially observed examples into a bounded-degree fragment of the sum-of-squares (SOS) hierarchy in polynomial time. Our algorithm performs two lifts simultaneously: (i) grounding-lift, where renaming-equivalent ground moments share one variable, collapsing the domain of individuals; and (ii) world-lift, where all pseudo-models (partial world assignm...
