Learning Novel Objects Continually Through Curiosity
This addresses the need for robots to learn continually in real-world environments, though it appears incremental as it builds on existing state-of-the-art methods.
The paper tackles the problem of enabling robots to learn unknown concepts continually by asking questions, similar to children's curiosity-driven learning. The results show that their curiosity-driven approach outperforms random sampling and softmax-based uncertainty sampling in classification accuracy and the total number of classes learned on a benchmark dataset.
Children learn continually by asking questions about the concepts they are most curious about. With robots becoming an integral part of our society, they must also learn unknown concepts continually by asking humans questions. The paper analyzes a recent state-of-the-art approach for continual learning. The paper further develops a self-supervised technique to find most of the uncertain objects in an environment by utilizing the cluster representation of the previously learned classes. We test our approach on a benchmark dataset for continual learning on robots. Our results show that our curiosity-driven continual learning approach beats random sampling and softmax-based uncertainty sampling in terms of classification accuracy and the total number of classes learned.