Abstract
Continual learning is essential for all real-world applications, as frozen pre-trained models cannot effectively deal with non-stationary data distributions. The purpose of this study is to review the state-of-the-art methods that allow continuous learning of computational models over time. We primarily focus on the learning algorithms that perform continuous learning in an online fashion from considerably large (or infinite) sequential data and require substantially low computational and memory resources. We critically analyze the key challenges associated with continual learning for autonomous real-world systems and compare current methods in terms of computations, memory, and network/model complexity. We also briefly describe the implementations of continuous learning algorithms under three main autonomous systems, i.e., self-driving vehicles, unmanned aerial vehicles, and urban robots. The learning methods of these autonomous systems and their strengths and limitations are extensively explored in this article.
Original language | English (US) |
---|---|
Article number | 9 |
Journal | Journal of Intelligent and Robotic Systems: Theory and Applications |
Volume | 105 |
Issue number | 1 |
DOIs | |
State | Published - May 2022 |
Keywords
- Autonomous systems
- Computations
- Continual learning
- Deep neural network
- Image recognition
- Memory
- Online learning
- Real-time
- Real-world
- Reinforcement learning
- Self-driving cars
- Supervised learning
- Unmanned aerial vehicles
- Urban robots
ASJC Scopus subject areas
- Software
- Control and Systems Engineering
- Mechanical Engineering
- Industrial and Manufacturing Engineering
- Electrical and Electronic Engineering
- Artificial Intelligence