Low-Shot Learning with Imprinted Weights

Hu, Qi; Brown, Matthew A.; Lowe, David

doi:10.1109/cvpr.2018.00610

Cited by 500 publications

(443 citation statements)

References 16 publications

Supporting

Mentioning

439

Contrasting

Order By: Relevance

“…Unfortunately, the preliminary results were disappointing, as we saw no significant improvements in accuracy when compared to using only the directly synthesized speech data. This parallels the findings of [10] where their augmentation of the small amount of data available for a new class failed to improve its classifier performance. We also suspect this may be due to the embedding model already having learned to deal with these distortions.…”

Section: Synthesized Speech Datasupporting

confidence: 76%

Training Keyword Spotters with Limited and Synthesized Speech Data

Lin

Kilgour

Roblek

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

With the rise of low power speech-enabled devices, there is a growing demand to quickly produce models for recognizing arbitrary sets of keywords. As with many machine learning tasks, one of the most challenging parts in the model creation process is obtaining a sufficient amount of training data. In this paper, we explore the effectiveness of synthesized speech data in training small, spoken term detection models of around 400k parameters. Instead of training such models directly on the audio or low level features such as MFCCs, we use a pre-trained speech embedding model trained to extract useful features for keyword spotting models. Using this speech embedding, we show that a model which detects 10 keywords when trained on only synthetic speech is equivalent to a model trained on over 500 real examples. We also show that a model without our speech embeddings would need to be trained on over 4000 real examples to reach the same accuracy.

show abstract

Section: Synthesized Speech Datasupporting

confidence: 76%

Training Keyword Spotters with Limited and Synthesized Speech Data

Lin

Kilgour

Roblek

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…We take pre-trained ResNet18 [5] with ImageNet as the feature extractor f ϕ . Train/test split setting is followed the suggestion of Imprinted Weights [47]. Here, 100 novel classes are required to be 33 distinguished, which is very challenging and similar to the real-world scenario.…”

Section: Caltech-ucsd Birdsmentioning

confidence: 99%

Few-Shot Learning for Domain-Specific Fine-Grained Image Classification

Sun

Dong

et al. 2021

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

Learning to recognize novel visual categories from a few examples is a challenging task for machines in realworld applications. In contrast, humans have the ability to discriminate even similar objects with little supervision. This paper attempts to address the few-shot fine-grained recognition problem. We propose a feature fusion model to explore the largest discriminative features by focusing on key regions. The model utilizes focus-area location to discover the perceptually similar regions among objects. High-order integration is employed to capture the interaction information among intra-parts. We also design a Center Neighbor Loss to form robust embedding space distribution for generating discriminative features. Furthermore, we build a typical fine-grained and few-shot learning dataset miniPPlankton from the real-world application in the area of marine ecological environment. Extensive experiments are carried out to validate the performance of our model. First the model is evaluated with two challenging experiments based on the miniDogsNet and Caltech-UCSD public datasets. The results demonstrate that our model achieves competitive performance compared with state-of-the-art models. Then, we implement our model for the real-world phytoplankton recognition task. The experimental results show the superiority of the proposed model compared with others on the miniPPlankton dataset.

show abstract

“…An interesting work regarding CNN classifiers using low-shot learning is given in [28]. The idea is to enable a model to successfully classify a newly seen category after being presented with merely few training examples.…”

Section: Future Workmentioning

confidence: 99%

“…Combining this work and DeepMimic might be very interesting, in the following sense. While using a mentor model trained on specific categories, upon the arrival of a novel category it might be easier to implant the new category in a student model combining the two processes described in DeepMimic and [28]. It is possible that a student model would adjust more naturally to new categories during the training process itself rather than an already trained model.…”

Section: Future Workmentioning

confidence: 99%

DeepMimic: Mentor-Student Unlabeled Data Based Training

Mosafi¹,

David²,

Netanyahu³

2019

Artificial Neural Networks and Machine Learning – ICANN 2019: Workshop and Special Sessions

View full text Add to dashboard Cite

In this paper, we present a deep neural network (DNN) training approach called the "DeepMimic" training method. Enormous amounts of data are available nowadays for training usage. Yet, only a tiny portion of these data is manually labeled, whereas almost all of the data are unlabeled. The training approach presented utilizes, in a most simplified manner, the unlabeled data to the fullest, in order to achieve remarkable (classification) results. Our DeepMimic method uses a small portion of labeled data and a large amount of unlabeled data for the training process, as expected in a real-world scenario. It consists of a mentor model and a student model. Employing a mentor model trained on a small portion of the labeled data and then feeding it only with unlabeled data, we show how to obtain a (simplified) student model that reaches the same accuracy and loss as the mentor model, on the same test set, without using any of the original data labels in the training of the student model. Our experiments demonstrate that even on challenging classification tasks the student network architecture can be simplified significantly with a minor influence on the performance, i.e., we need not even know the original network architecture of the mentor. In addition, the time required for training the student model to reach the mentor's performance level is shorter, as a result of a simplified architecture and more available data. The proposed method highlights the disadvantages of regular supervised training and demonstrates the benefits of a less traditional training approach.

show abstract

Low-Shot Learning with Imprinted Weights

Cited by 500 publications

References 16 publications

Training Keyword Spotters with Limited and Synthesized Speech Data

Training Keyword Spotters with Limited and Synthesized Speech Data

Few-Shot Learning for Domain-Specific Fine-Grained Image Classification

DeepMimic: Mentor-Student Unlabeled Data Based Training

Contact Info

Product

Resources

About