A meta-learning approach towards microvessel classification based on PAC-Bayes

In this work, we proposed a meta-learning method for the classification of microvessel images based on PAC-Bayes. We first introduce the modeling of Single-Opponent (SO) neurons to capture the color information of microvessel images. Then, we presented the PAC-Bayes bound on multiple learning tasks for the classification of microvessel images by optimizing the PAC-Bayes objective function. Further, we summarize the meta-learning algorithm based on PACBayes to classify microvessel images in detail. The proposed method is superior in precision and f1-score compared with other representative methods.

[1]  Masanori Tanaka,et al.  Microvessel Morphology and Vascular Endothelial Growth Factor Expression in Human Colonic Carcinoma With or Without Metastasis , 2002, Laboratory Investigation.

[2]  Thomas Serre,et al.  A New Biologically Inspired Color Image Descriptor , 2012, ECCV.

[3]  Ambedkar Dukkipati,et al.  Generative Adversarial Residual Pairwise Networks for One Shot Learning , 2017, ArXiv.

[4]  Pieter Abbeel,et al.  Meta-Learning with Temporal Convolutions , 2017, ArXiv.

[5]  A. El‐Naggar,et al.  Expression levels of genes that regulate metastasis and angiogenesis correlate with advanced pathological stage of renal cell carcinoma. , 2001, The American journal of pathology.

[6]  David A. McAllester.  PAC-Bayesian model averaging , 1999, COLT '99.

[7]  Hugo Larochelle,et al.  Optimization as a Model for Few-Shot Learning , 2016, ICLR.

[8]  Ajit Rajwade,et al.  Block and Group Regularized Sparse Modeling for Dictionary Learning , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Gregory R. Koch,et al.  Siamese Neural Networks for One-Shot Image Recognition , 2015 .

[10]  Umit Topaloglu,et al.  Automatic identification of angiogenesis in double stained images of liver tissue , 2009, BMC Bioinformatics.

[11]  Quan Wen,et al.  Biologically inspired classification of microvessel histopathology via sparse coding , 2013, 2013 Sixth International Conference on Advanced Computational Intelligence (ICACI).

[12]  Juan Chen,et al.  A predictive coding approach on microvessel identification via Single-Opponent signals , 2013, 2013 6th International Congress on Image and Signal Processing (CISP).

[13]  Richard S. Zemel,et al.  Prototypical Networks for Few-shot Learning , 2017, NIPS.

[14]  R. Shapley,et al.  Color in the Cortex: single- and double-opponent cells , 2011, Vision Research.

[15]  Sergey Levine,et al.  Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks , 2017, ICML.

[16]  P. V. Van Schil,et al.  Growth index is independent of microvessel density in non-small cell lung carcinomas. , 2003, Human pathology.

[17]  Paolo Frasconi,et al.  Bilevel Programming for Hyperparameter Optimization and Meta-Learning , 2018, ICML.

[18]  Oriol Vinyals,et al.  Matching Networks for One Shot Learning , 2016, NIPS.

[19]  Ron Meir,et al.  Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theory , 2017, ICML.

[20]  Hong Yu,et al.  Meta Networks , 2017, ICML.