DAI505 Advanced Deep Machine Learning (A/651/0609) DAI505 Assignment Brief Qualification Level 5 Diploma in Artificial Intelligence (610/3935/4) Unit Code DAI505 Unit Title Advanced Deep Machine Learning Unit

DAI505 Advanced Deep Machine Learning (A/651/0609) DAI505 Assignment Brief Qualification Level 5 Diploma in Artificial Intelligence (610/3935/4) Unit Code DAI505 Unit Title Advanced Deep Machine Learning Unit Reference A/651/0609 Credits 20 TQT 200 GLH 120 Assignment Aim unit builds on the fundamentals of deep learning, data sciences, machine learning and Neural networks covered in the Level 4-unit Introduction to Deep Learning (AID405). Students will explore advanced Neural network architecture and techniques and their use in improving the performance of deep learning models. Students will have the opportunity to use deep learning models within NLP and Reinforcement Learning on given real-life business scenarios.

Learning Outcomes and Assignment Criteria Learning Outcomes When awarded credit for this unit, a learner will:

Assessment Criteria Assessment of this learning outcome will require a learner to demonstrate that they can:

1. Be able to review contemporary theories and concepts of deep machine learning.  

1.1 Summarize contemporary theories, concepts and models of deep machine learning. 1.2 Analyse the effectiveness of various deep machine learning models in addressing different business scenarios. 2. Understand advance Neural network architecture.  

2.1 Investigate the use of advance Neural network architectures 2.2 Explain the principles governing transfer learning. 2.3 Explain the principles of improving trainedmodels 3. Be able to use different techniques to improve performance of deep machine learning models. 3.1 Investigate the hyperparameter tuning and generative models to improve performance. 3.2 Use hyperparameter tuning to review impacts from different parameters on model. 3.3 Use generative models to review effects on image generations and data synthesis. 4. Use deep machine learning models on Natural Language Processing (NLP) and Reinforced Machine Learning (RML). 4.1 Apply deep learning techniques in NLP tasks and examine the results. 4.2 Integrate deep learning techniques within RML algorithms to solve complex problems.