MLx Fundamentals 25

This module provides a comprehensive overview of machine learning and deep learning fundamentals, focusing on supervised learning techniques, neural network architectures, and the role of feature mapping. Students will explore autoencoders, causal inference, and the limitations of neural networks, equipping them with both theoretical knowledge and practical skills essential for interpreting and applying machine learning models effectively.

1. 1. Talk
2. 2. Q&A

This module delves into the essential principles of artificial intelligence, emphasizing self-supervised learning and the architecture of neural networks. Students will explore the evolution of deep learning, the significance of large language models, and the impact of data on model performance. By understanding these concepts, learners will gain insights into optimizing AI models for various applications.

1. 1. Talk
2. 2. Q&A

In this module, students will delve into essential machine learning techniques, focusing on clustering with k-means, classification using naive Bayes, and optimization methods like gradient descent. Through practical applications in Google Colab, learners will gain insights into data pattern discovery, predictive modeling, and the significance of optimization in enhancing model performance.

1. 1. Talk

This module delves into the essential optimization algorithms that drive machine learning success. Students will learn about gradient descent, the differences between convex and non-convex optimization, and advanced methods like accelerated and stochastic gradient descent. By understanding these concepts, learners will be equipped to tackle challenges in noisy environments and improve model performance effectively.

1. 1. Talk

This module delves into the intricacies of generative models, focusing on their applications in image and video generation. Students will explore the evolution of text-to-video technologies, the significance of model architecture, and the challenges of training efficiency. By understanding the differences between generative and discriminative models, as well as the role of conditioning and evaluation metrics, learners will gain valuable insights into the future of AI-generated content.

1. 1. Generative modelling
2. 2. Video generation
3. 3. Q&A

In this module, students will explore the foundational concepts of deep learning, focusing on the practical implementation of neural networks. They will learn to build and train models using NumPy, apply them to real-world datasets like MNIST and CIFAR, and gain insights into critical components such as activation functions and backpropagation. This knowledge is essential for anyone looking to advance in the field of machine learning.

1. 1. Talk

This module delves into the evolution and applications of generative AI, focusing on image, video, and 3D generation technologies. Students will explore diffusion models, continual learning frameworks, and techniques to maintain consistency in generative outputs. By understanding the interplay between model architecture, training methodologies, and data quality, learners will gain valuable insights into the future of AI-driven content creation.

1. 1. Talk
2. 2. Q&A

In this module, students will explore the foundational concepts of diffusion models, focusing on their ability to generate realistic images from random noise. Through practical examples and hands-on implementation using Python, learners will gain a deep understanding of the forward and reverse processes, model architecture, and the critical denoising steps involved in training and inference.

1. 1. Talk

This module delves into the fundamentals of generative AI and natural language processing, covering the architecture of language models, fine-tuning techniques, and the integration of local data. Students will learn how to effectively train and adapt large language models for specific tasks, ensuring optimal performance while preserving knowledge. The insights gained will empower students to leverage generative AI tools in real-world applications.

1. 1. Introduction to Generative AI for NLP
2. 2. Q&A
3. 3. Expanding the LLM knowledge
4. 4. Q&A

In this module, students will explore the training and inference processes of large language models (LLMs) through hands-on applications. They will learn to implement sentiment analysis, zero-shot classification, and text generation using the Transformers library, while also gaining experience with APIs from leading companies like OpenAI. This knowledge is essential for leveraging LLMs in real-world scenarios.

1. 1. Talk