Malik Payton AI 4364

This week focused on Large Language Models (LLMs), which are one of the most advanced forms of artificial intelligence today. These models, like ChatGPT, are designed to understand and generate human-like text using a transformer-based architecture. I learned that LLMs go through two major stages: pre-training on massive datasets and fine-tuning for specific tasks. This process allows them to perform a wide range of functions, from answering questions to generating content. In Google Colab, I explored how to use a transformer model to summarize large pieces of text. This exercise showed me how AI can quickly take complex information and turn it into something concise and useful. It made me think about how businesses could use this for things like reports, marketing content, customer service, and even social media management. Seeing this in action helped me understand why LLMs are becoming such an important tool across industries. This week really tied everything together from the cou...

This week focused on comparing two of the most popular AI frameworks: TensorFlow and PyTorch. I learned that both tools are extremely powerful, but they are used differently depending on the situation. TensorFlow is known for being strong in production environments and large-scale deployment, while PyTorch is more commonly used for research and experimentation. Understanding this difference helped me realize that choosing the right tool is just as important as understanding AI itself. One of the biggest takeaways for me was how PyTorch uses dynamic computation graphs, which makes it easier to test and debug code. This aligns with my experience so far, as PyTorch feels more intuitive and beginner-friendly. On the other hand, TensorFlow has a larger ecosystem and is often used by companies for real-world applications because of its scalability. Learning about these differences gave me a better perspective on how AI projects move from development to real-world use. This week also helped...

This week focused on PyTorch and how to actually build and train machine learning models. I learned that PyTorch is a powerful and flexible library that allows developers to create neural networks using tensors, which are essentially multi-dimensional arrays. One thing that stood out to me is how beginner-friendly PyTorch feels compared to other tools, especially because of its clear structure and ability to debug easily. In Google Colab, I followed along with building a neural network that predicts housing prices using the Boston Housing dataset. This process helped me understand the full workflow of training a model, including splitting the data, running multiple epochs, calculating loss, and improving predictions over time. Learning about concepts like Mean Squared Error (MSE) and Stochastic Gradient Descent (SGD) gave me a better understanding of how models actually improve. At first, the terminology was a little overwhelming, but breaking it down step-by-step made it much more ma...

This week focused on Natural Language Processing (NLP), which is how AI understands and works with human language. I learned about text classification, sentiment analysis, and language modeling. One of the biggest takeaways was how AI can analyze emotions in text, like determining whether a review is positive or negative. This is something businesses can use to understand customer feedback at scale. I also explored how models like ChatGPT work by predicting the next word in a sentence. That helped me understand why AI responses feel so natural. Running NLP examples in Google Colab showed how text is tokenized and processed before being analyzed. It made me realize how much work goes on behind the scenes when we interact with AI tools daily. This week was probably the most relevant to my field. As a marketing major, NLP can be used for social media analysis, brand sentiment tracking, and customer engagement. It showed me how companies can turn text data into real business insights, wh...

This week focused on deep learning and neural networks, which are more advanced forms of machine learning. I learned that neural networks are inspired by the human brain and use layers of “neurons” to process information. Concepts like activation functions (ReLU, sigmoid, softmax) showed me how models decide what information matters. It was interesting to see how these systems can handle complex tasks like image recognition and speech processing. In Google Colab, I followed along with building a neural network to classify handwritten digits. Even though the code looked complex at first, breaking it down step-by-step made it manageable. When I saw the model reach around 97% accuracy, it really clicked for me how powerful these systems are. This made me think about real-world uses like facial recognition or automated systems in business. This week showed me how deep learning goes beyond basic predictions and can handle more complex data like images and language. As someone interested i...

This week introduced me to the foundation of artificial intelligence: machine learning. I learned that machine learning allows computers to identify patterns in data and make decisions without being directly programmed. There are three main types: supervised learning, unsupervised learning, and reinforcement learning. Supervised learning stood out the most because it uses labeled data to predict outcomes, which made it easier for me to understand compared to the others. I also worked in Google Colab to run a simple linear regression model. Even though I’m not a coding expert, I was able to follow along and understand what the code was doing by asking ChatGPT to explain each part. Seeing the model generate predictions and display a regression line made everything feel more real and practical. It helped me realize that AI isn’t just theory—it’s something I can actually use. Overall, this week built my confidence with AI. Instead of being overwhelmed, I started to see how these tools co...