TY - GEN
T1 - Learning Permutation Invariant Representations Using Memory Networks
AU - Kalra, Shivam
AU - Adnan, Mohammed
AU - Taylor, Graham
AU - Tizhoosh, H. R.
N1 - Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Many real-world tasks such as classification of digital histopathology images and 3D object detection involve learning from a set of instances. In these cases, only a group of instances or a set, collectively, contains meaningful information and therefore only the sets have labels, and not individual data instances. In this work, we present a permutation invariant neural network called Memory-based Exchangeable Model (MEM) for learning universal set functions. The MEM model consists of memory units that embed an input sequence to high-level features enabling it to learn inter-dependencies among instances through a self-attention mechanism. We evaluated the learning ability of MEM on various toy datasets, point cloud classification, and classification of whole slide images (WSIs) into two subtypes of the lung cancer—Lung Adenocarcinoma, and Lung Squamous Cell Carcinoma. We systematically extracted patches from WSIs of the lung, downloaded from The Cancer Genome Atlas (TCGA) dataset, the largest public repository of WSIs, achieving a competitive accuracy of 84.84% for classification of two sub-types of lung cancer. The results on other datasets are promising as well, and demonstrate the efficacy of our model.
AB - Many real-world tasks such as classification of digital histopathology images and 3D object detection involve learning from a set of instances. In these cases, only a group of instances or a set, collectively, contains meaningful information and therefore only the sets have labels, and not individual data instances. In this work, we present a permutation invariant neural network called Memory-based Exchangeable Model (MEM) for learning universal set functions. The MEM model consists of memory units that embed an input sequence to high-level features enabling it to learn inter-dependencies among instances through a self-attention mechanism. We evaluated the learning ability of MEM on various toy datasets, point cloud classification, and classification of whole slide images (WSIs) into two subtypes of the lung cancer—Lung Adenocarcinoma, and Lung Squamous Cell Carcinoma. We systematically extracted patches from WSIs of the lung, downloaded from The Cancer Genome Atlas (TCGA) dataset, the largest public repository of WSIs, achieving a competitive accuracy of 84.84% for classification of two sub-types of lung cancer. The results on other datasets are promising as well, and demonstrate the efficacy of our model.
KW - Medical images
KW - Multi Instance Learning
KW - Permutation invariant models
KW - Whole slide image classification
UR - http://www.scopus.com/inward/record.url?scp=85093111443&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85093111443&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-58526-6_40
DO - 10.1007/978-3-030-58526-6_40
M3 - Conference contribution
AN - SCOPUS:85093111443
SN - 9783030585259
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 677
EP - 693
BT - Computer Vision – ECCV 2020 - 16th European Conference, Proceedings
A2 - Vedaldi, Andrea
A2 - Bischof, Horst
A2 - Brox, Thomas
A2 - Frahm, Jan-Michael
PB - Springer Science and Business Media Deutschland GmbH
T2 - 16th European Conference on Computer Vision, ECCV 2020
Y2 - 23 August 2020 through 28 August 2020
ER -