Generating Classification Weights With GNN Denoising Autoencoders for Few-Shot Learning

By Spyros Gidaris, Nikos Komodakis. 2019

244 citations recorded by the source corpus.

Given an initial recognition model already trained on a set of base classes, the goal of this work is to develop a meta-model for few-shot learning. The meta-model, given as input some novel classes with few training examples per class, must properly adapt the existing recognition model into a new model that can correctly classify in a unified way both the novel and the base classes. To accomplish this goal it must learn to output the appropriate classification weight vectors for those two types of classes. To build our meta-model we make use of two main innovations: we propose the use of a Denoising Autoencoder network (DAE) that (during training) takes as input a set of classification weights corrupted with Gaussian noise and learns to reconstruct the target-discriminative classification weights. In this case, the injected noise on the classification weights serves the role of regularizing the weight generating meta-model. Furthermore, in order to capture the co-dependencies between different classes in a given task instance of our meta-model, we propose to implement the DAE model as a Graph Neural Network (GNN). In order to verify the efficacy of our approach, we extensively evaluate it on ImageNet based few-shot benchmarks and we report state-of-the-art results.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Generalized and Incremental Few-Shot Learning by Explicit Learning and Calibration without Forgetting (2021, methodology): “The technique is widely used to avoid the explicit calibration of magnitudes [19, 37, 40, 31, 15, 36].”
Self-Promoted Prototype Refinement for Few-Shot Class-Incremental Learning (2021, background): “[9] introduces Graph Neural Network to capture the co-dependencies between adjacent classes and promote the weight generation of new classes.”
Incremental Few-Shot Object Detection (2020, neutral): “Note that there are many more recent FSL methods [6, 26, 7, 17, 47, 55, 50, 27, 39, 11, 16] that produce better performance on image classification tasks than MAML.”
Boosting Few-Shot Visual Learning With Self-Supervision (2019, neutral): “, training images, of a few-shot task [26, 33, 53, 55, 57, 59]; methods learning to map a test example to a class label by accessing memory modules that store training examples for that task [11, 25, 35, 38, 50]; approaches that learn how to generate the weights of a classifier [12, 16, 43, 44, 13] ”
Interventional Few-Shot Learning (2020, methodology): “For example, φ can be the classifier weight generator [19], distance kernel function in kNN [62], or even θ’s initialization [18].”
Variational Few-Shot Learning (2019, background): “[15, 36, 48, 16, 3] further implicitly adopt meta-learning as an auxiliary phase to predict the parameters from the activations in the last stage.”
DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learning (2020, background): “classification problem, such as works based on graph theories [66], [67], [68], [69], reinforcement learning [70], differentiable SVM [71], generative models [72], [73], [74], [75], [76], [77], [78], [79], [80], transductive learning [81], [82], [83], [84], [85], recurrent models [86], [87], self-su”
Few-Shot Object Detection With Attention-RPN and Multi-Relation Detector (2019, background): “In [43, 44] the authors introduced Graph Neural Network (GNN) to model relationship between different categories.”
Adaptive Subspaces for Few-Shot Learning (2020, unknown): “12 wDAE-GNN [54] WRN-28-10 62.”
DeepEMD: Few-Shot Image Classification With Differentiable Earth Mover’s Distance and Structured Classifiers (2020, unknown): “Apart from the two popular branches, many other promising methods are also proposed to handle the few-shot classification problem, such as works based on graph theories [13, 15, 21], reinforcement learning [6], differentiable”
Incremental few-shot learning via vector quantization in deep embedded space (2021, methodology): “These methods are referred to as dynamic few-shot learning (Gidaris & Komodakis, 2018; Ren et al., 2019a; Gidaris & Komodakis, 2019).”
Improved Few-Shot Visual Classification (2019, unknown): “Few-shot learning approaches typically take one of two forms: 1) nearest neighbor approaches and their variants, including matching networks [40], which effectively apply nearest-neighbor or weighted nearest neighbor classification on the samples themselves, either in a feature [15, 16, 34] or a sem”
Cross Attention Network for Few-shot Classification (2019, methodology): “Following [40, 35, 34, 19, 9, 14, 7], we adopt the episode training mechanism, which has been demonstrated as an effective approach for few-shot learning.”
A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, methodology): “At present, graph neural networks are widely used for tasks such as few-shot image classification [145], [146], semantic segmentation [147] and instance segmentation tasks.”
Relational Generalized Few-Shot Learning (2019, methodology): “Their work is extended in [15] by a graph neural network (GNN) based denoising autoencoder to regularize the class prototypes, where the underlying graph is a special case of our fully connected setup.”