文獻翻譯
- 人工智能
- 《Meta - Learning with Memory - Augmented Neural Networks》
- one-shot learning：
- Neural Turing Machines，NTMs
- 《Model - Agnostic Meta - Learning for Fast Adaptation of Deep Networks》
- Meta - learning
- gradient steps
- finetune
- 《Attention Is All You Need》
- 《Imagenet Classification with Deep Convolutional Neural Networks》
- 《Automatic Chain of Thought Prompting in Large Language Models》
人工智能
《Meta - Learning with Memory - Augmented Neural Networks》
https://proceedings.mlr.press/v48/santoro16.pdf
文獻綜述：
Despite recent breakthroughs in the applications of deep neural networks, one setting that presents a persistent challenge is that of “one-shot learning.” Traditional gradient-based networks require a lot of data to learn, often through extensive iterative training. When new data is encountered,the models must inefficiently relearn their parameters to adequately incorporate the new information without catastrophic interference. Architectures with augmented memory capacities, such as Neural Turing Machines (NTMs), offer the ability to quickly encode and retrieve new information, and hence can potentially obviate the downsides of conventional models. Here, we demonstrate the ability of a memory-augmented neural network to rapidly assimilate new data, and leverage this data to make accurate predictions after only a few samples. We also introduce a new method for accessing an external memory that focuses on memory content, unlike previous methods that additionally use memory location-based focusing mechanisms.
Abstract
Despite recent breakthroughs（突破） in the applications of deep neural networks（深度神經網絡）, one setting（一種背景） that presents（提出） a persistent（持久的；堅持不懈的；持續不斷的） challenge is that of “one-shot learning.（一次性學習）” Traditional gradient-based（基于梯度的） networks require a lot of data to learn, often through extensive（廣泛的） iterative（迭代的） training. When new data is encountered（遇到）,the models must inefficiently relearn their parameters to adequately incorporate（吸收） the new information without （不在。。情況下）catastrophic（災難的） interference（干擾）. Architectures（架構） with **augmented （增強的）**memory capacities（性能）, such as Neural Turing Machines (NTMs)（神經圖靈機）, offer the ability to quickly encode and retrieve（檢索） new information, and hence（因此） can potentially（潛在地） obviate（消除；排除；使不必要） the downsides（缺點） of conventional（傳統的） models. Here, we demonstrate（證明） the ability of a memory-augmented neural network to rapidly（迅速地） assimilate（同化） new data, and leverage（杠桿） this data to make accurate（準確的） predictions after only a few samples. We also introduc
