Lifelong Zero-Shot Learning

By Kun-Juan Wei, Cheng Deng, Xu Yang. 2020

74 citations recorded by the source corpus.

Zero-Shot Learning (ZSL) handles the problem that some testing classes never appear in training set. Existing ZSL methods are designed for learning from a fixed training set, which do not have the ability to capture and accumulate the knowledge of multiple training sets, causing them infeasible to many real-world applications. In this paper, we propose a new ZSL setting, named as Lifelong Zero-Shot Learning (LZSL), which aims to accumulate the knowledge during the learning from multiple datasets and recognize unseen classes of all trained datasets. Besides, a novel method is conducted to realize LZSL, which effectively alleviates the Catastrophic Forgetting in the continuous training process. Specifically, considering those datasets containing different semantic embeddings, we utilize Variational Auto-Encoder to obtain unified semantic representations. Then, we leverage selective retraining strategy to preserve the trained weights of previous tasks and avoid negative transfer when fine-tuning the entire model. Finally, knowledge distillation is employed to transfer knowledge from previous training stages to current stage. We also design the LZSL evaluation protocol and the challenging benchmarks. Extensive experiments on these benchmarks indicate that our method tackles LZSL problem effectively, while existing ZSL methods fail.

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

1. Yanan Gu
2. Cheng Deng
3. Kun-Juan Wei
4. Dacheng Tao
5. Tongliang Liu
6. Gan Sun
7. Xu Yang
8. Pravendra Singh

In Computer Vision and Pattern Recognition

1. Cheng Deng
2. Tongliang Liu
3. Yanan Gu
4. Kun-Juan Wei
5. Dacheng Tao
6. Xu Yang
7. Gan Sun
8. Piyush Rai

In Radiology, Nuclear Medicine and Imaging

1. Yanan Gu
2. Cheng Deng
3. Kun-Juan Wei
4. Dacheng Tao
5. Pravendra Singh
6. Vinay P. Namboodiri
7. Piyush Rai
8. Xu Yang

Citations in the corpus

• Class-Incremental Instance Segmentation via Multi-Teacher Networks (2021, methodology): “Specifically, we utilize knowledge distillation to simulate the guidance process, which is proposed in (Hinton, Vinyals, and Dean 2015) and widely used in incremental tasks (Li and Hoiem 2017; Hou et al. 2018; Wei, Deng, and Yang 2020) and other tasks (Yang et al. 2020c).”
• Rectification-based Knowledge Retention for Continual Learning (2021, methodology): “For the task incremental generalized zero-shot learning problem, we use the setting described in [30].”
• Continual Representation Learning for Biometric Identification (2020, neutral): “Continual learning is also named lifelong learning [46,56,58,67], incremental learning [6,49,55] and sequential learning [5, 9] in previous works.”
• Heterogeneous Graph Attention Network for Unsupervised Multiple-Target Domain Adaptation (2020, background): “In most cases, these domains have similar underlying data distributions, which can be leveraged to build more effective and robust model for the task of across multiple domains [12].”
• Not Just Selection, but Exploration: Online Class-Incremental Continual Learning via Dual View Consistency (2022, background): “Continual learning [21, 32, 35, 37, 51, 52] aims to build a model that can learn sequentially and accumulate acquired knowledge over time.”
• A Survey on Few-Shot Class-Incremental Learning (2023, background): “Additionally, in order to tackle the issue of the inability to learn from data streams in ZSL, Wei et al. (2020, 2021) have proposed the concept of incremental zero-shot learning (IZSL).”
• I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting (2020, background): “…in great quantifies of research fields, e.g., autonomous driving (Behl et al. 2017; Li et al. 2020), intelligent robotics (Stria and Hlavc 2018; Wei, Deng, and Yang 2020; Chen et al. 2020), object clustering (Yang et al. 2020; Zhang et al. 2020a, 2021), medical diagnosis (Dong et al. 2019,…”