Generative Adversarial Residual Pairwise Networks for One Shot Learning

By Akshay Mehrotra, Ambedkar Dukkipati. 2017

118 citations recorded by the source corpus.

Deep neural networks achieve unprecedented performance levels over many tasks and scale well with large quantities of data, but performance in the low-data regime and tasks like one shot learning still lags behind. While recent work suggests many hypotheses from better optimization to more complicated network structures, in this work we hypothesize that having a learnable and more expressive similarity objective is an essential missing component. Towards overcoming that, we propose a network design inspired by deep residual networks that allows the efficient computation of this more expressive pairwise similarity objective. Further, we argue that regularization is key in learning with small amounts of data, and propose an additional generator network based on the Generative Adversarial Networks where the discriminator is our residual pairwise network. This provides a strong regularizer by leveraging the generated data samples. The proposed model can generate plausible variations of exemplars over unseen classes and outperforms strong discriminative baselines for few shot classification tasks. Notably, our residual pairwise network design outperforms previous state-of-theart on the challenging mini-Imagenet dataset for one shot learning by getting over 55% accuracy for the 5-way classification task over unseen classes.

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

Data Augmentation Generative Adversarial Networks (2017, methodology): “…use a hierarchical Boltzmann machine, through to modern deep learning architectures for one-shot conditional generation in (Rezende et al., 2016), hierarchical variational autoencoders in (Edwards & Storkey, 2017) and most recently a GAN based one-shot generative model (Mehrotra & Dukkipati, 2017).”
Few-Shot Learning with Graph Neural Networks (2017, neutral): “Pair-Wise Mehrotra & Dukkipati (2017) - - 94.8% - Prototypical Networks Snell et al. (2017) 97.4% 99.3% 95.4% 98.8% ConvNet with Memory Kaiser et al. (2017) 98.4% 99.6% 95.0% 98.6% Agnostic Meta-learner Finn et al. (2017) 98.7 ± 0.4% 99.9 ± 0.3% 95.8 ± 0.3% 98.9 ± 0.2% Meta Networks Munkhdalai & Yu…”
Meta-Transfer Learning for Few-Shot Learning (2018, methodology): “Several methods propose to learn a data generator e.g. conditioned on Gaussian noise [29, 44, 54].”
Learning to Self-Train for Semi-Supervised Few-Shot Classification (2019, background): “They are based on either data augmentation [15, 29] or gradient descent [3, 23, 7, 5, 42, 19, 27, 32, 10].”
A Closer Look At Feature Space Data Augmentation For Few-Shot Intent Classification (2019, methodology): “For example, generative models are used for data augmentation in image classification (Mehrotra and Dukkipati, 2017; Antoniou et al., 2018; Zhang et al., 2018), text classification (Gupta, 2019), anomaly detection (Lim et al.”
A Two-Stage Approach to Few-Shot Learning for Image Recognition (2019, methodology): “Some ad-hoc approaches to address few-shot learning were to carry out data augmentation by harnessing unlabeled data [12], by transformation and adding noise [13], [14], and by synthesizing artiﬁcial examples [15]– [18] or using compositional representations [19], [20].”
Collect and Select: Semantic Alignment Metric Learning for Few-Shot Learning (2019, background): “regard the generated fake instances as a strong regularizer [24].”
Dynamic Input Structure and Network Assembly for Few-Shot Learning (2017, methodology): “Other architectures such as pairwise networks demonstrated by Mehrotra and Dukkipati (2017) have been used successfully as well.”
DADA: Deep Adversarial Data Augmentation for Extremely Low Data Regime Classification (2018, background): “Training a deep learning classiﬁer on small datasets is a topic of wide interests in the ﬁelds of ﬁne-grained visual recognition [15], few-shot learning [16], and life-long learning in new environments [17].”
Improving One-Shot Learning through Fusing Side Information (2017, background): “Other recent work, including Skip Residual Pairwise Net (SRPN) (Mehrotra and Dukkipati, 2017) and Prototypical Networks (Snell et al.”
Meta-Learning Requires Meta-Augmentation (2020, background): “Mehrotra and Dukkipati [23] train a generator to generate new examples for one-shot learning problems, and Liu et al. [22] applied task augmentation to few-shot image classiﬁcation by rotating images, deﬁning each rotation as a new task.”
A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, unknown): “One is to use generative models to caste prior knowledge [13], [14], [15], where bayesian programming learning [16] is the most representative framework [17] in this field.”
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”