Model zoo

By T. Beardsley. 1984

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

DeepInspect: A Black-box Trojan Detection and Mitigation Framework for Deep Neural Networks (2019, background): “Caffe Model Zoo 1 is an example platform where pre-trained models are publicly shared with the users.”
Differentially Private Model Publishing for Deep Learning (2019, background): “A large number of pre-trained models have been publicly available in model zoo repositories [7].”
SCNN: An accelerator for compressed-sparse convolutional neural networks (2017, methodology): “Table II lists the attributes of three commonly used networks in image processing: AlexNet [20], GoogLeNet [28], and VGGNet [27], whose speciﬁcations come from the Caffe BVLC Model Zoo [4].”
Cloud-Based or On-Device: An Empirical Study of Mobile Deep Inference (2017, methodology): “Because Caffe provides a flexible way to define CNN layers and has a large collection of pre-trained models [14], we choose to evaluate CNN models trained using the Caffe framework in this paper.”
Backdoor attacks against learning systems (2017, background): “With a large collection of PLMs available for off-the-shelf use (e.g., Model Zoo [9]), it comes as no surprise that a non-computer science major student built a full-ﬂedged search engine in less than three weeks [10].”