Evidential Learning and Reasoning (ELR)

Evidential Learning and Reasoning aims at providing uncertainty-aware learning and reasoning capabilities to AI systems, so to encompass both epistemic and aleatory uncertainty.

For further details, please see the following tutorials:

• ELP-tutorial

Achievements

ELR features in

• Enabling rapidly formed human-agent coalition teams through extensible information exchange;
• Uncertainty-Aware AI&ML.

It comprises research both in Evidential Deep Learning (EDL) and Evidential Logic Programming (ELP).

Publications

Relevant ELR publications are:

• Murat Sensoy, Lance Kaplan, Melih Kandemir, Evidential Deep Learning to Quantify Classification Uncertainty, NeurIPS 2018, 2018
• Federico Cerutti, Lance Kaplan, Angelika Kimmig, Murat Sensoy, Probabilistic Logic Programming with Beta-Distributed Random Variables, AAAI 2019, 2019.
• Murat Sensoy, Lance Kaplan, Federico Cerutti, Maryam Saleki, Uncertainty-Aware Deep Classifiers using Generative Models, AAAI 2020, 2020
• Lance Kaplan, Federico Cerutti, Murat Sensoy, Kumar Vijay Mishra, Second-Order Learning and Inference using Incomplete Data for Uncertain Bayesian Networks: A Two Node Example, Fusion 2020, 2020.
• Federico Cerutti, Lance Kaplan, Angelika Kimmig, Murat Sensoy, Handling Epistemic and Aleatory Uncertainties in Probabilistic Circuits, Submitted, TBD.