Evidential Turing Processes

Melih Kandemir, Abdullah Akgül, Manuel Haussmann, Gozde Unal

Research output: Contribution to conferencePaperScientificpeer-review

2 Citations (Scopus)

Abstract

A probabilistic classifier with reliable predictive uncertainties i) fits successfully to the target domain data, ii) provides calibrated class probabilities in difficult regions of the target domain (e.g. class overlap), and iii) accurately identifies queries coming out of the target domain and rejects them. We introduce an original combination of Evidential Deep Learning, Neural Processes, and Neural Turing Machines capable of providing all three essential properties mentioned above for total uncertainty quantification. We observe our method on five classification tasks to be the only one that can excel all three aspects of total calibration with a single standalone predictor. Our unified solution delivers an implementation-friendly and compute efficient recipe for safety clearance and provides intellectual economy to an investigation of algorithmic roots of epistemic awareness in deep neural nets.

Original languageEnglish
Pages1-16
Number of pages16
Publication statusPublished - 2022
MoE publication typeNot Eligible
EventInternational Conference on Learning Representations - Virtual, Online
Duration: 25 Apr 202229 Apr 2022
Conference number: 10

Conference

ConferenceInternational Conference on Learning Representations
Abbreviated titleICLR
CityVirtual, Online
Period25/04/202229/04/2022

Fingerprint

Dive into the research topics of 'Evidential Turing Processes'. Together they form a unique fingerprint.

Cite this