eprintid: 10067254 rev_number: 18 eprint_status: archive userid: 608 dir: disk0/10/06/72/54 datestamp: 2019-02-07 18:03:55 lastmod: 2021-09-20 22:36:51 status_changed: 2019-02-07 18:03:55 type: article metadata_visibility: show creators_name: Kittler, J creators_name: Zor, C title: Delta Divergence: A Novel Decision Cognizant Measure of Classifier Incongruence ispublished: pub divisions: UCL divisions: B04 divisions: C05 divisions: F48 keywords: Classifier incongruence, divergence clutter, f-divergences, total variation distance note: This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/ abstract: In pattern recognition, disagreement between two classifiers regarding the predicted class membership of an observation can be indicative of an anomaly and its nuance. Since, in general, classifiers base their decisions on class a posteriori probabilities, the most natural approach to detecting classifier incongruence is to use divergence. However, existing divergences are not particularly suitable to gauge classifier incongruence. In this paper, we postulate the properties that a divergence measure should satisfy and propose a novel divergence measure, referred to as delta divergence. In contrast to existing measures, it focuses on the dominant (most probable) hypotheses and, thus, reduces the effect of the probability mass distributed over the non dominant hypotheses (clutter). The proposed measure satisfies other important properties, such as symmetry, and independence of classifier confidence. The relationship of the proposed divergence to some baseline measures, and its superiority, is shown experimentally. date: 2019-06 date_type: published official_url: http://doi.org/10.1109/TCYB.2018.2825353 oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1622593 doi: 10.1109/TCYB.2018.2825353 lyricists_name: Zor, Cemre lyricists_id: CZORX93 actors_name: Flynn, Bernadette actors_id: BFFLY94 actors_role: owner full_text_status: public publication: IEEE Transactions on Cybernetics volume: 49 number: 6 pagerange: 2331-2343 event_location: United States issn: 2168-2275 citation: Kittler, J; Zor, C; (2019) Delta Divergence: A Novel Decision Cognizant Measure of Classifier Incongruence. IEEE Transactions on Cybernetics , 49 (6) pp. 2331-2343. 10.1109/TCYB.2018.2825353 <https://doi.org/10.1109/TCYB.2018.2825353>. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10067254/1/08370682.pdf