Updated version: Clarified the contribution (see abstract, intro, and conclusion); added figures to illustrate the architecture of the network and the difference between training and generation; different selection of experiments in Section 6.4; some textual edits
Deep learning research over the past years has shown that by increasing the scope or difficulty of the learning problem over time, increasingly complex learning problems can be addressed. We study incremental learning in the context of sequence learning, using generative RNNs in the form of multi-layer recurrent Mixture Density Networks. While the potential of incremental or curriculum learning to enhance learning is known, indiscriminate application of the principle does not necessarily lead to improvement, and it is essential therefore to know which forms of incremental or curriculum learning have a positive effect. This research contributes to that aim by comparing three instantiations of incremental or curriculum learning.
We introduce Incremental Sequence Learning, a simple incremental approach to sequence learning. Incremental Sequence Learning starts out by using only the first few steps of each sequence as training data. Each time a performance criterion has been reached, the length of the parts of the sequences used for training is increased.
We introduce and make available a novel sequence learning task and data set: predicting and classifying MNIST pen stroke sequences. We find that Incremental Sequence Learning greatly speeds up sequence learning and reaches the best test performance level of regular sequence learning 20 times faster, reduces the test error by 74%, and in general performs more robustly; it displays lower variance and achieves sustained progress after all three comparison methods have stopped improving. The other instantiations of curriculum learning do not result in any noticeable improvement. A trained sequence prediction model is also used in transfer learning to the task of sequence classification, where it is found that transfer learning realizes improved classification performance compared to methods that learn to classify from scratch.
Curriculum-DPO++: Direct Preference Optimization via Data and Model Curricula for Text-to-Image Generation
Department of Computer Science, University of Bucharest(布加勒斯特大学计算机科学系);University of Trento(特伦托大学);Center for Research in Computer Vision (CRCV), University of Central Florida(中央佛罗里达大学计算机视觉研究中心)
On the Benefit of Optimal Transport for Curriculum Reinforcement Learning
Technical University of Darmstadt(达姆施塔特工业大学);German Research Center for AI (DFKI)(德国人工智能研究中心);Centre for Cognitive Science at Technical University of Darmstadt(达姆施塔特工业大学认知科学中心);Center for Artificial Intelligence and Data Science at University of Würzburg(维尔茨堡大学人工智能与数据科学中心);Aalto University(阿尔托大学)
Seoul National University(首尔国立大学);Automation and Systems Research Institute(自动化与系统研究所);Artificial Intelligence Institute of Seoul National University(首尔国立大学人工智能研究所)
The Lie of the Average: How Class Incremental Learning Evaluation Deceives You?
School of Artificial Intelligence, Nanjing University(人工智能学院,南京大学);National Key Laboratory for Novel Software Technology, Nanjing University(新型软件技术国家实验室,南京大学);School of Computing and Artificial Intelligence, Southwestern University of Finance and Economics(计算与人工智能学院,西南财经大学)
Curriculum Learning for Reinforcement Learning Domains: A Framework and Survey
University of Texas at Austin(德克萨斯大学奥斯汀分校);University of Oxford(牛津大学);University of Leeds(利兹大学);Tufts University(塔夫茨大学);Alberta Machine Intelligence Institute(阿尔伯塔机器智能研究所);University of Alberta(阿尔伯塔大学);Sony AI(索尼人工智能)
AdaHAT: Adaptive Hard Attention to the Task in Task-Incremental Learning
School of Mathematical Sciences, Peking University(北京大学数学科学学院);Population Health Sciences Institute, Newcastle University(纽卡斯尔大学人口健康科学研究所);School of Engineering Mathematics and Technology, University of Bristol(布里斯托尔大学工程数学与技术学院);School of Computing and Creative Technologies, University of the West of England(西英格兰大学计算与创意技术学院)
Long-term simulation of physical and mechanical behaviors using curriculum-transfer-learning based physics-informed neural networks
College of Mechanics and Engineering Science, Hohai University(河海大学力学与工程科学学院);China Three Gorges Corporation(中国长江三峡集团有限公司);Cairo University(开罗大学);Leibniz Universität Hannover(汉诺威莱布尼茨大学)