Today, I continued my AI journey with the University of Helsinki’s Elements of AI course. As always, ChatGPT remains my trusted study buddy, helping me debate, analyze, and challenge ideas rather than just memorizing facts.
📚 Exploring AI's Related Fields
This section introduced key disciplines related to AI, including:
✅ Machine Learning – Systems that improve with experience and data.
✅ Deep Learning – A subset of ML focused on complex neural networks.
✅ Data Science – A broad field incorporating AI, ML, statistics, and business applications.
✅ Robotics – AI applied to real-world autonomous machines.
The first exercise tested my understanding of how these fields overlap. I had to correctly categorize AI, ML, DL, Data Science, and Computer Science in a Venn Diagram-style taxonomy exercise.
📌 Final Score: 5/5! 🎯 Nailed it!
🤖 The AI Debate – When Logic and Course Answers Don’t Always Align
The second exercise required classifying AI applications into statistics, robotics, or machine learning. My initial choices were solid, but ChatGPT and I debated some of them—leading to some "incorrect" answers according to the course.
📌 Final Score: 2/5. However, if I had stuck to my original choices, I would have scored 4/5.
Now, here’s the important part:
Even though ChatGPT’s suggestions lowered my score, I wouldn’t change a thing.
💡 Why? Because the debate itself was where I learned the most.
Two key discussions stood out:
1️⃣ Gallup Polling (Statistics vs. ML) – My original answer was only statistics, but ChatGPT suggested adding machine learning. The course said ML was wrong, but based on real-world applications, I can still see how AI models could be used to analyze Gallup data beyond just statistics.
2️⃣ Missile Guidance (Robotics vs. Statistics) – I initially chose only robotics, but ChatGPT suggested adding statistics. The course marked it incorrect, saying guidance is mostly a physics problem, but ChatGPT’s reasoning made sense—there are statistical models used in trajectory prediction.
In both cases, the course followed a strict classification, but real-world applications are rarely so rigid. The discussions themselves helped me understand the nuances of AI far better than just memorizing the “correct” answers.
🎯 Key Takeaways from Today
✅ Scored 5/5 in Exercise 2, proving a solid grasp of AI's related fields.
✅ AI is hard to define, but its key characteristics are autonomy and adaptivity.
✅ Machine learning is a subfield of AI, while deep learning is a subfield of ML.
✅ Robotics and AI often go hand in hand, but not all automated systems are true robotics.
✅ Debating AI concepts is more valuable than simply memorizing answers.
✅ Course exercises are useful, but real-world AI applications often blur the lines between categories.
No comments:
Post a Comment