Few-shot and meta-learning methods for image understanding: a survey

He, Kai; Pu, Nan; Lao, Mingrui; Lew, Michael S.

doi:10.1007/s13735-023-00279-4

Cited by 8 publications

(3 citation statements)

References 124 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meta-learning, or "learning to learn," aims to develop models that can learn new skills or adapt quickly to new environments with minimal training instances. There are three general approaches: 1) learning efficient distance metrics (metric-based); 2) using iterative networks with external or internal memory (model-based); and 3) optimizing model parameters explicitly for fast learning (optimization-based) [11].…”

Section: Meta-learning Approach To Improve Metacognition and Creativitymentioning

confidence: 99%

Enhancing Metacognitive and Creativity Skills through AI-Driven Meta-Learning Strategies

Khusnul Khotimah,

Rusijono,

Andi Mariono

2024

Int. J. Interact. Mob. Technol.

View full text Add to dashboard Cite

This study investigates the efficacy of a meta-learning approach in improving metacognitive and creative skills. This quantitative study focused on an experimental group using a onegroup pretest-posttest research design. All participants underwent a pretest to assess their initial metacognitive abilities and were subsequently exposed to a meta-learning framework throughout the course. A post-test was conducted to assess the impact of the intervention. The findings indicate a statistically significant improvement in metacognitive skills from the pretest to the post-test. This study confirms the effectiveness of meta-learning strategies and elucidates the relationship between meta-learning and metacognition. Meta-learning enables students to comprehend their own learning processes, thereby improving their capacity to strategize, oversee, and control their cognitive functions with the assistance of artificial intelligence (AI). This approach incorporates creative elements that can stimulate metacognitive thinking, encouraging students to adjust their learning strategies and think outside the box. This research suggests that meta-learning can improve metacognitive abilities, providing valuable insights into educational technology and course design in higher education settings.

show abstract

Section: Meta-learning Approach To Improve Metacognition and Creativitymentioning

confidence: 99%

Enhancing Metacognitive and Creativity Skills through AI-Driven Meta-Learning Strategies

Khusnul Khotimah,

Rusijono,

Andi Mariono

2024

Int. J. Interact. Mob. Technol.

View full text Add to dashboard Cite

show abstract

“…Meta-learning (Vanschoren, 2018;Elsken et al, 2020;Li et al, 2021;Liu et al, 2022;He et al, 2023;Vettoruzzo et al, 2024) was put forward to solve the problem of few-shot learning. It empowers learning systems with the ability to acquire knowledge from multiple tasks, enabling faster adaptation and generalization to new tasks (Vettoruzzo et al, 2024).…”

Section: Introductionmentioning

confidence: 99%

The meta-learning method for the ensemble model based on situational meta-task

Zhang,

Zhou,

et al. 2024

Front. Neurorobot.

View full text Add to dashboard Cite

IntroductionThe meta-learning methods have been widely used to solve the problem of few-shot learning. Generally, meta-learners are trained on a variety of tasks and then generalized to novel tasks.MethodsHowever, existing meta-learning methods do not consider the relationship between meta-tasks and novel tasks during the meta-training period, so that initial models of the meta-learner provide less useful meta-knowledge for the novel tasks. This leads to a weak generalization ability on novel tasks. Meanwhile, different initial models contain different meta-knowledge, which leads to certain differences in the learning effect of novel tasks during the meta-testing period. Therefore, this article puts forward a meta-optimization method based on situational meta-task construction and cooperation of multiple initial models. First, during the meta-training period, a method of constructing situational meta-task is proposed, and the selected candidate task sets provide more effective meta-knowledge for novel tasks. Then, during the meta-testing period, an ensemble model method based on meta-optimization is proposed to minimize the loss of inter-model cooperation in prediction, so that multiple models cooperation can realize the learning of novel tasks.ResultsThe above-mentioned methods are applied to popular few-shot character datasets and image recognition datasets. Furthermore, the experiment results indicate that the proposed method achieves good effects in few-shot classification tasks.DiscussionIn future work, we will extend our methods to provide more generalized and useful meta-knowledge to the model during the meta-training period when the novel few-shot tasks are completely invisible.

show abstract

“…Currently, deep learning is always tightly related to big data. It has become a unanimous agreement that the success of deep learning is the result of the big labeled datasets [1,2]. Deep learning provides a kind of end-to-end approach for object recognition that is more generalizable and robust than the traditional methods based on feature extraction, but it is difficult to implement because of its need for large annotated datasets [3].…”

Section: Introductionmentioning

confidence: 99%

Local Contrast Learning for One-Shot Learning

Zhang,

Yuan,

Luo

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Learning a deep model from small data is an opening and challenging problem. In high-dimensional spaces, few samples only occupy an extremely small portion of the space, often exhibiting sparsity issues. Classifying in this globally sparse sample space poses significant challenges. However, by using a single sample category as a reference object for comparing and recognizing other samples, it is possible to construct a local space. Conducting contrastive learning in this local space can overcome the sparsity issue of a few samples. Based on this insight, we proposed a novel deep learning approach named Local Contrast Learning (LCL). This is analogous to a key insight into human cognitive behavior, where humans identify the objects in a specific context by contrasting them with the objects in that context or from their memory. LCL is used to train a deep model that can contrast the recognized sample with a couple of contrastive samples that are randomly drawn and shuffled. On a one-shot classification task on Omniglot, the deep model-based LCL with 86 layers and 1.94 million parameters, which was trained on a tiny dataset with only 60 classes and 20 samples per class, achieved an accuracy of 98.95%. Furthermore, it achieved an accuracy of 99.24% at 156 classes and 20 samples per class. LCL is a fundamental idea that can be applied to alleviate the parametric model’s overfitting resulting from a lack of training samples.

show abstract

Few-shot and meta-learning methods for image understanding: a survey

Cited by 8 publications

References 124 publications

Enhancing Metacognitive and Creativity Skills through AI-Driven Meta-Learning Strategies

Enhancing Metacognitive and Creativity Skills through AI-Driven Meta-Learning Strategies

The meta-learning method for the ensemble model based on situational meta-task

Local Contrast Learning for One-Shot Learning

Contact Info

Product

Resources

About