#5, First Floor, 4th Street , Dr. Subbarayan Nagar, Kodambakkam, Chennai-600 024 pro@slogix.in

Office Address

  • #5, First Floor, 4th Street Dr. Subbarayan Nagar Kodambakkam, Chennai-600 024 Landmark : Samiyar Madam
  • pro@slogix.in
  • +91- 81240 01111

Social List

Transfer Learning

Transfer Learning is a supervised learning method that refers to improving the performance of the target domain by transferring the knowledge to different source domains. Transfer learning utilizes the gained knowledge from solving a particular problem for different problems related to that particular problem. The primary significance of transfer learning is that learning from scratch of a considerable amount of data is no longer needed and uses less computational sources and less time due to its pre-learned knowledge from different domains. Transfer learning is categorized into three types involving Transductive learning, inductive learning, unsupervised transfer learning, and negative learning. Transductive transfer learning refers to the label information from the source domain. If the label information of the target-domain instances is available, it belongs to inductive transfer learning. If the label information is unknown for both the source and the target domains, it uses unsupervised transfer learning. The negative learning transfers unrelated knowledge from the source domain. Transfer learning approaches for each type are categorized into four such as instance-based, feature-based, parameter-based, and relational-based approaches. Transfer learning for machine learning involves pre-trained models reused in machine learning models and provides the efficient model with deployment for multi models. Recently, Deep learning in transfer learning is the growing technique due to training due to less data for training deep neural networks. The general application of transfer learning is cancer subtype discovery, building utilization, general game playing, text classification, digit recognition, medical imaging, bio-informatics, intelligent transport system, a recommendation system, and spam filtering. The future research area of transfer learning is transfer learning with user privacy, sentiment analysis, micro-expression recognition, voice, and gesture recognition.