A Generative Model For Zero Shot Learning Using Conditional Variational Autoencoders

Mishra, Alok; Reddy, M Shiva Krishna; Mittal, Anurag; Murthy, Hema A.

doi:10.48550/arxiv.1709.00663

Cited by 12 publications

(24 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For AWA, the preferred range is k ∈ [7,17] and the most preferred k is around 12. For CUB, the preferred range is k ∈ [4,19] and the most preferred k is around 16. Secondly, we evaluate the training time.…”

Section: Results and Analysismentioning

confidence: 99%

Adaptive Adjustment with Semantic Feature Space for Zero-shot Recognition

Guo

2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In most recent years, zero-shot recognition (ZSR) has gained increasing attention in machine learning and image processing fields. It aims at recognizing unseen class instances with knowledge transferred from seen classes. This is typically achieved by exploiting a pre-defined semantic feature space (FS), i.e., semantic attributes or word vectors, as a bridge to transfer knowledge between seen and unseen classes. However, due to the absence of unseen classes during training, the conventional ZSR easily suffers from domain shift and hubness problems. In this paper, we propose a novel ZSR learning framework that can handle these two issues well by adaptively adjusting semantic FS. To the best of our knowledge, our work is the first to consider the adaptive adjustment of semantic FS in ZSR. Moreover, our solution can be formulated to a more efficient framework that significantly boosts the training. Extensive experiments show the remarkable performance improvement of our model compared with other existing methods.

show abstract

Section: Results and Analysismentioning

confidence: 99%

Adaptive Adjustment with Semantic Feature Space for Zero-shot Recognition

Guo

2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Multi-label Zero-shot Learning is a less explored problem compared to single-label zero-shot learning [10,20,24,26,32]. Fu et al [7] enumerate all possible combinations of labels and thus transform the problem into single-label zero-shot learning.…”

Section: Related Workmentioning

confidence: 99%

“…Without the yellow nodes from ImageNet[23], it would be hard to predict the unseen class Sheep since other seen classes are not semantically related to it. However, with the knowledge that the current image has a high probability on the ImageNet class Bighorn Sheep and the word embeddings of both Sheep and Bighorn Sheep, the model can better predict the unseen class by referring to the ImageNet classes.Although zero-shot learning for single-label classification has been well studied[10,20,24,26,32,33], multi-label zero-shot learning (ML-ZSL)[7,14,18,35] is a less explored area. An image is associated with potentially multiple seen and unseen classes, but only labels of seen classes are provided during training.…”

mentioning

confidence: 99%

Multi-label Zero-shot Classification by Learning to Transfer from External Knowledge

Huang

Chen

Tang

et al. 2020

Preprint

View full text Add to dashboard Cite

Multi-label zero-shot classification aims to predict multiple unseen class labels for an input image. It is more challenging than its single-label counterpart. On one hand, the unconstrained number of labels assigned to each image makes the model more easily overfit to those seen classes. On the other hand, there is a large semantic gap between seen and unseen classes in the existing multi-label classification datasets. To address these difficult issues, this paper introduces a novel multi-label zero-shot classification framework by learning to transfer from external knowledge. We observe that ImageNet is commonly used to pretrain the feature extractor and has a large and fine-grained label space. This motivates us to exploit it as external knowledge to bridge the seen and unseen classes and promote generalization. Specifically, we construct a knowledge graph including not only classes from the target dataset but also those from ImageNet. Since ImageNet labels are not available in the target dataset, we propose a novel PosVAE module to infer their initial states in the extended knowledge graph. Then we design a relational graph convolutional network (RGCN) to propagate information among classes and achieve knowledge transfer. Experimental results on two benchmark datasets demonstrate the effectiveness of the proposed approach. IntroductionRecent deep learning methods are able to outperform humans in image classification tasks by leveraging supervised learning and large-scale datasets such as ImageNet [23]. However,

show abstract

“…At test time, the label for an unseen-class test input is the class that maximizes the VAE lower bound. Afterwards, Mishra et al (2017) trained a conditional VAE (Sohn et al, 2015) to learn the underlying probability distribution of the image features conditioned on the class embedding vector. Similarly, Arora et al (2018) proposed a method able to generate semantically rich CNN feature distributions via the conditional VAE with discriminator-driven feedback mechanism improving the reconstruction capability.…”

Section: Extension Matchingmentioning

confidence: 99%

Small Sample Learning in Big Data Era

Shu,

Xu,

Meng

2018

Preprint

View full text Add to dashboard Cite

As a promising area in artificial intelligence, a new learning paradigm, called Small Sample Learning (SSL), has been attracting prominent research attention in the recent years. In this paper, we aim to present a survey to comprehensively introduce the current techniques proposed on this topic. Specifically, current SSL techniques can be mainly divided into two categories. The first category of SSL approaches can be called "concept learning", which emphasizes learning new concepts from only few related observations. The purpose is mainly to simulate human learning behaviors like recognition, generation, imagination, synthesis and analysis. The second category is called "experience learning", which usually co-exists with the large sample learning manner of conventional machine learning. This category mainly focuses on learning with insufficient samples, and can also be called small data learning in some literatures. More extensive surveys on both categories of SSL techniques are introduced and some neuroscience evidences are provided to clarify the rationality of the entire SSL regime, and the relationship with human learning process. Some discussions on the main challenges and possible future research directions along this line are also presented.

show abstract

A Generative Model For Zero Shot Learning Using Conditional Variational Autoencoders

Cited by 12 publications

References 0 publications

Adaptive Adjustment with Semantic Feature Space for Zero-shot Recognition

Adaptive Adjustment with Semantic Feature Space for Zero-shot Recognition

Multi-label Zero-shot Classification by Learning to Transfer from External Knowledge

Small Sample Learning in Big Data Era

Contact Info

Product

Resources

About