Open Access Open Access  Restricted Access Subscription Access

Meta-Learning: Unleashing the Power of Self-Improving Artificial Intelligent (AI) Systems

Sunny Orike, Donald S. Ene

Abstract


Meta-learning, or "learning to learn," is an emerging field within artificial intelligence (AI) that enables AI systems to acquire knowledge and adapt to new tasks. This paper explores the concept of meta-learning and its applications in enabling self-improving AI systems. It delves into key components and methodologies involved in meta-learning, discusses various approaches such as model-agnostic meta-learning (MAML) and recurrent neural networks (RNNs); and examines applications, challenges, and ethical considerations. Through comprehensive analysis and review of existing literature, this paper aims to inspire further advancements in self-improving AI systems.


Full Text:

PDF

References


Antoniou, A., Storkey, A., & Edwards, H., 2018. How to train your MAML. Inter. Conf. on Learning Representations (ICLR).

Bertinetto, L., Henriques, J. F., Valmadre, J., Torr, P. H., Vedaldi, A., 2019. Meta-Learning with differentiable convex optimization. Proc. of IEEE Conf. on Comp. Vision and Pattern Recog. 76, 3-12.

Bolukbasi, T., Chang, K. W., Zou, J. Y., Saligrama, V., Kalai, A. T., 2016. Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Adv. in Neu. Infor. Proc. Sys. 29 (43),49-57.

Finn, C., Abbeel, P., Levine, S., 2017. Model-agnostic meta-learning for fast adaptation of deep networks. Proc. of the 34th Inter. Conf. on Machine Learning (ICML), 1126-1135.

Finn, C., Yu, T., Zhang, T., Abbeel, P., Levine, S., 2017b. One-Shot Visual Imitation Learning via Meta-Learning. Proc. of the 35th Inter. Conf. on Machine Learning, 70, 2428-2437.

Gu, J., Lu, Z., Li, H., Li, V. O., 2018. Meta-transfer learning for few-shot learning. Proc. of the 32nd AAAI Conference on Artificial Intelligence.

Hochreiter, S., Schmidhuber, J., 1997. Long short-term memory. Neural Comp. 9 (8), 1735-1780.

Kim, B., Khanna, R., & Koyejo, O. O., 2020. MATE: meta-analysis for transparent explanations. Adv. in Neural Infor. Processing Sys. 13, 433-445.

Lake, B. M., Salakhutdinov, R., and Tenenbaum, J. B., 2015. Human-level concept learning through probabilistic program induction. Science 350(6266),13, 32-38.

Li, L., Jamieson, K. G., Rostamizadeh, A., Gonina, E., 2018. Hyperband: a novel bandit-based approach to hyperparameter optimization. J. Machine Learning R. 18, 1-52.


Refbacks

  • There are currently no refbacks.