Transferrable Prototypical Networks for Unsupervised Domain Adaptation

Abstract

In this paper, we introduce a new idea for unsupervised domain adaptation via a remold of Prototypical Networks, which learn an embedding space and perform classification via a remold of the distances to the prototype of each class. Specifically, we present Transferrable Prototypical Networks (TPN) for adaptation such that the prototypes for each class in source and target domains are close in the embedding space and the score distributions predicted by prototypes separately on source and target data are similar. Technically, TPN initially matches each target example to the nearest prototype in the source domain and assigns an example a "pseudo" label. The prototype of each class could then be computed on source-only, target-only and source-target data, respectively. The optimization of TPN is end-to-end trained by jointly minimizing the distance across the prototypes on three types of data and KLdivergence of score distributions output by each pair of the prototypes. Extensive experiments are conducted on the transfers across MNIST, USPS and SVHN datasets, and superior results are reported when comparing to state-of-theart approaches. More remarkably, we obtain an accuracy of 80.4% of single model on VisDA 2017 dataset.

Joint Domain Alignment and Discriminative Feature Learning for Unsupervised Deep Domain Adaptation

“…Compared with the semantic domain alignment methods (TPN [12], DIAL [14], SCA [15]) and the target classifier learning methods (iCAN [17]), our method outperforms them in most transfer tasks by jointly optimizing semantic domain alignment and target classifier learning in the feature space. Compared with state-of-the-art methods, SDA-TCL achieves better or comparable performance in all transfer tasks.…”

“…On par with these methods aligning distributions in the feature space, some methods align distributions in raw pixel space by translating source data to the target domain with Image to Image translation techniques [25][26][27][28][29][30]. In addition to domain-level distribution alignment, the class-level information in target data is also frequently used to align class-level distributions [11][12][13][14][15]31]. Compared with these methods, our method not only aligns class-level distributions, but also learns target discriminative features.…”

“…Baseline Methods. We compared our proposed SDA-TCL with state-of-the-art methods: (I) Domainlevel distribution alignment methods: Gradient Reversal (RevGrad) [9], Similarity Learning (SimNet) [20]; (II) Class-level distribution alignment methods: Transferable Prototypical Networks (TPN) [12], Domain-Invariant Adversarial Learning (DIAL) [14], Similarity Constrained Alignment (SCA) [15]; (III) Aligning distributions on pixel-level methods: Coupled Generative Adversarial Network (CoGAN) [25], Cycle-Consistent Adversarial Domain Adaptation (CyCADA) [29], Generate To Adapt (GTA) [30]; (IV) Utilizing pseudo-labels implicitly methods: Maximum Classifier Discrepancy (MCD) [35]; (V)Learning target classifier methods: Incremental Collaborative and Adversarial Network (iCAN) [17]; (V)State-of-the-art Methods: Joint Adaptation Network (JAN) [18], Deep adversarial Attention Alignment (DAAA) [22], Self-Ensembling (S-En) [38], Conditional Domain Adversarial Networks (CDAN+E) [24]. With the same protocol, we cite the results from the papers respectively following the previous methods [11,16].…”

Focus on Semantic Consistency for Cross-Domain Crowd Understanding

“…Therefore, a well-trained model on the GCC dataset doesn't work well in the real world. Some of the recent work [17][18][19][20][21] provided us with a domain adaptation strategy, using image style transfer networks to narrow the domain gap. For the cross-domain crowd counting problem, [16] also proposed a domain adaptation method to make synthetic data closer to real data in visual perception via the SE Cycle GAN [16] network.…”

Novel Few-shots Learning Neural Network for Predicting Carbohydrate-active Enzyme (CAZyme) Affinity Towards Fructo-oligosaccharides

“…A neural network optimization algorithm of finding the loss of each round of neural network generation is a core feature of few-shot learning (Garcia, V. and Joan B., 2017). The loss algorithm applied is based on a prototypical neural network with adjustment of using accuracy rate instead of Euclidean distance; since among the various neural networks available, a prototypical neural network is the most reliable means of approach in this situation thanks to its outstanding performance in the small sample space in practices (Pan, Y. et al, 2019), which often outputs the prediction accuracy that has surpassed human recognition (He, K. et al, 2016). By integrating with few-shot learning algorithms, the prototypical neural network achieved an approximated 70% accuracy in 5-way 5-shot image classification (Richard, Z. et al, 2017).…”