one year of studying (it was a mistake)

Summary

In this reflective video, the speaker shares a year-long journey of learning math, computer science, and AI, all while juggling a full-time job and a new baby. While they gained valuable knowledge, they highlight the importance of applying what they learned through hands-on projects rather than just consuming theoretical content. They discuss their scattered approach to learning, which left them with no tangible results, and explain how focusing on project-based learning would have been more effective. The speaker aims to share their future project-based learning journey in upcoming videos, encouraging viewers to engage with their process.

Takeaways

• 😀 **Mathematics is fundamental**: Understanding math, especially linear algebra and discrete math, is crucial for AI, but it requires dedicated effort to master.
• 😀 **Self-learning can be overwhelming**: Trying to cover too many topics at once can lead to burnout and a lack of real progress. Focus is key.
• 😀 **Practical application is essential**: Consuming theoretical knowledge without applying it through projects or exercises leads to shallow learning.
• 😀 **Too much time spent on irrelevant topics**: Spending time on CUDA programming or other areas not aligned with your immediate goals can be a distraction.
• 😀 **Project-based learning is the most effective**: Starting with a small, manageable project allows you to apply new knowledge, reinforcing your learning in a practical way.
• 😀 **Courses alone are not enough**: Watching videos or reading books can provide basic understanding, but you need to engage in hands-on exercises to really internalize concepts.
• 😀 **Avoiding rabbit holes is crucial**: While deep dives are necessary in certain fields (like math), it’s important not to get lost in overly complex areas that don’t serve your core objectives.
• 😀 **Focus on a single goal**: Instead of jumping between unrelated topics, choosing a focused project can drive your learning and give it structure.
• 😀 **Algorithms and data structures are key**: Solidifying your understanding of algorithms and practicing through platforms like LeetCode helps improve problem-solving skills.
• 😀 **Tools and frameworks evolve quickly**: Staying up-to-date with emerging tools (like PyTorch) is essential for staying relevant in the AI and machine learning space.

Q & A

Why did the speaker regret spending so much time on math and computer science?

The speaker felt that while studying math and computer science was intellectually interesting, it wasn't the best use of their time for their career goals. They realized that much of the advanced math and computer science knowledge they acquired wouldn't directly apply to their daily work, especially in software engineering and AI.

What was the initial motivation behind studying AI and machine learning?

The speaker was bored with web development and sought a new challenge. AI seemed like an exciting field to explore, which led them to pick up books and start studying the subject.

What did the speaker learn about their preparation for AI and machine learning?

The speaker discovered that they were not adequately prepared for the advanced level of math required in AI and machine learning. For example, they struggled with textbooks like 'Deep Learning' and found that they needed more foundational knowledge in math and programming tools like Python libraries (Pandas, Numpy).

How did the speaker approach learning math for AI and machine learning?

The speaker started by watching lecture series on linear algebra and discrete math and later used Math Academy's 'Mathematics for Machine Learning' course. They emphasized that merely watching videos or reading about math isn't enough—you need to actively practice to build a deep understanding.

What did the speaker find frustrating about watching math lectures?

The speaker found that watching math lectures, like those by 3Blue1Brown, could give you a good understanding of the concepts but didn't equip them to solve problems. To truly learn, they had to apply what they learned through practice.

What does the speaker consider the most effective way to solidify knowledge in subjects like math and algorithms?

The speaker believes that the best way to solidify knowledge is through active practice, such as doing exercises and solving problems. This approach helps build a deeper understanding, as opposed to passive learning from lectures or reading.

Why did the speaker study computer science topics like low-level hardware and algorithms?

The speaker studied computer science topics because they were interested in areas like CUDA programming and algorithms. However, they realized that diving into these topics without focusing on practical application (like programming in C++) was not the most efficient approach.

What mistake did the speaker make when studying computer science and algorithms?

The speaker invested time in learning low-level concepts like CUDA programming without having a clear direction or immediate need for them. They realized that they should have focused more on practical skills, like learning C++ for CUDA kernel programming, if they were serious about pursuing that path.

How did the speaker’s focus shift over the year of studying AI and related fields?

Initially, the speaker was scattered across multiple areas, including math, computer science, algorithms, and machine learning. They later realized that without a focused project or goal, they were not making meaningful progress in any one area.

What would the speaker do differently if they could restart their learning journey?

If they could start over, the speaker would focus on completing smaller, project-based tasks that could be accomplished within a month. These projects would stretch their skills but also help them consolidate knowledge by applying it in practical scenarios. This focused approach would allow them to build momentum and avoid getting lost in irrelevant rabbit holes.