Given the course is from 2017, Is still worth taking the course? or is the majority of concepts outdated?
As someone who took it this year, I will say that it is super worth taking and it will be a good use of your time.
Here are a few reasons:
- You begin from the basics. Implementing neural networks using numpy from scratch.
- The core concepts related to deep learning are explained in detail, including optimization algorithms and hyperparameter tuning.
- The course contents, especially the labs usually get updated to reflect some advancements
- The basics of deep learning are still the same. To deeply understand it, you need to understand the concepts in depth. For instance, convolution is still the same, LSTMs and GRUs are the way they are. Sure, things have modernized, but the way they work (which was explained) remains the same.
- You can always try your hands on applying the concepts learned in practical projects in the real world, just to stay in touch.
The choice is yours ultimately. But the Deep Learning Specialization is still relevant especially for people interested in breaking into AI. As proof, here is something I lifted from huggingface’s NLP course intro:
Is better taken after an introductory deep learning course, such as ... or one of the programs developed by DeepLearning.AI
The course covers a wide range of topics. It’s a good survey of deep learning methods. They are all still relevant.
Thanks a lot for that.
I also completed the DLS within the last 6 months, and I found that they have done a good job updating the course based on the latest developments in AI. Not only were the fundamentals still relevant, Prof Andrew also gave interesting insights to the latest trends (within the last year) from time to time.
Got it. Thanks.