TY  - JOUR
A1  - Mundt, Martin
A1  - Pliushch, Iuliia
A1  - Majumder, Sagnik
A1  - Hong, Yongwon
A1  - Ramesh, Visvanathan
T1  - Unified probabilistic deep continual learning through generative replay and open set recognition
T2  - Journal of imaging
N2  - Modern deep neural networks are well known to be brittle in the face of unknown data instances and recognition of the latter remains a challenge. Although it is inevitable for continual-learning systems to encounter such unseen concepts, the corresponding literature appears to nonetheless focus primarily on alleviating catastrophic interference with learned representations. In this work, we introduce a probabilistic approach that connects these perspectives based on variational inference in a single deep autoencoder model. Specifically, we propose to bound the approximate posterior by fitting regions of high density on the basis of correctly classified data points. These bounds are shown to serve a dual purpose: unseen unknown out-of-distribution data can be distinguished from already trained known tasks towards robust application. Simultaneously, to retain already acquired knowledge, a generative replay process can be narrowed to strictly in-distribution samples, in order to significantly alleviate catastrophic interference.
KW  - catastrophic forgetting
KW  - continual deep learning
KW  - deep generative models
KW  - open-set recognition
KW  - variational inference
Y1  - 2022
UR  - http://publikationen.ub.uni-frankfurt.de/frontdoor/index/index/docId/82823
UR  - https://nbn-resolving.org/urn:nbn:de:hebis:30:3-828237
SN  - 2313-433X
N1  - Funding: EU H2020 Project AEROBI ; 687384
N1  - Funding: EU H2020 Project RESIST ; 769066
N1  - Funding: BMBF project AISEL ; 01IS19062
N1  - Additional financial support from Goethe University was instrumental in concluding the research.
VL  - 8
IS  - 4, art. 93
SP  - 1
EP  - 34
PB  - MDPI
CY  - Basel
ER  -