Start Learning Machine Learning II today!

Course Description

'Machine Learning II' is an advanced course designed as a progression from 'Machine Learning Foundations,' delving deeper into intricate aspects of neural network design and optimization. Students explore the construction of complex layers like the Inception model and the strategic utilization of 1x1 convolution layers in 3D as bottlenecks, honing their ability to craft sophisticated architectures while understanding parameter efficiency.

The curriculum extends beyond mere model construction, delving into practical applications such as configuring pre-existing models and fine-tuning them for specific tasks. Furthermore, it introduces innovative paradigms where image patches are treated as tokens, converted into embeddings, and augmented with positional encodings. Students also engage with attention mechanisms like attention dropout layers and self-attention, crucial for localizing objects and capturing global dependencies between class tokens.

By utilising theoretical foundations with hands-on applications, 'Machine Learning II' equips students with advanced techniques pivotal for creating state-of-the-art machine learning models. This course not only emphasizes the construction of intricate neural network architectures but also offers insights into methodologies essential for diverse tasks in image analysis, object localization, and leveraging attention mechanisms for enhanced model performance and understanding global contexts within data.