This site is meant as a repository for information on Lifelong Machine Learning and Reasoning, or LMLR. It has been developed and is maintained by the Machine Learning Research Group at Acadia University, Nova Scotia, Canada.
The Machine Learning Research Group (MLRG) undertakes research into novel machine learning and data mining algorithms and approaches, and the application of these approaches to synthetic and real-world problems. The lab’s researchers and students specialize in developing machine learning algorithms and methods particularly those in the area of Lifelong Machine Learning, Transfer Learning, Knowledge Consolidation, and Learning to Reason. The group has particular expertise in artificial neural networks and deep learning used for supervised, unsupervised and time series problems. We also apply standard machine learning methods as well as more advanced LMLR approaches to problems in the areas of data mining, adaptive systems, intelligent agents and robotics. For more information please see the Software, Data and Publication pages.
Researchers can submit links to their work to be added to this website or contribute datasets and software to help the collection grow. Please also see the ResearhGate LMLR Project at https://www.researchgate.net/project/Lifelong-Machine-Learning-and-Reasoning
For a nice overview and rational for LMLR, please see Lifelong machine learning systems: Beyond learning algorithms. Proceedings of the AAAI Spring Symposium on Lifelong Machine Learning by Silver, D. Yang, Q. and Li, AAAI, March, 2013, pp 49–55.
The following link is to an old but popular Introduction to Sequential Life Long Learning by D. Silver.
- 2015 – Lifelong Machine Learning and Reasoning, invited talk at the NIPS 2015 workshop on Cognitive Computation: Integrating Neural and Symbolic Approaches, Montreal, December 2015 – https://www.researchgate.net/publication/286920204_Lifelong_Machine_Learning_and_Reasoning
- 2014 – Unsupervised Deep Multimodal Learning system that can scale to many sensory / motor channels (Best Paper Award FLAIRS-2016) – https://ml3cpu.acadiau.ca/.
- 2013 – Consolidation using Sweep Task Rehearsal: Overcoming the Stability-Plasticity Problem – https://www.researchgate.net/publication/220343617_Inductive_transfer_with_context-sensitive_neural_networks
- 2010 – CsMTL MLP for WEKA: Neural network learning with inductive transfer (see Software section) – https://www.researchgate.net/publication/221437820_CsMTL_MLP_for_WEKA_Neural_network_learning_with_inductive_transfer
- 2005 – Context-sensitive Multiple Task Learning (csMTL) networks that can produce superior models to standard MTL approaches and provide can be used in Lifelong Machine Learning research and application – https://www.researchgate.net/publication/220343617_Inductive_transfer_with_context-sensitive_neural_networks