what must be the right order to start with? One should learn deep learning first or it must be machine learning? Between somewhere lies Artificial Intelligence and Linux... So what must be right the order to grab the basics of the all described above.
Honestly, the best way to learn ML and DL is to start with mathematics. You must know very well algebra, vectors & matrices, have a decent knowledge of calculus, and most importantly know very well statistics and probability.
Do you know that the very first developed ML technique is well over one hundred years old? It is the linear regression that was developed within statistics.
Only when you have this knowledge, it is suitable to start with ML. I recommend you to use Python to write down programs as there are available very good libraries of ML techniques.
The last thing would be deep learning with its neuronal networks.
Lastly, I highly recommend you to study complex systems as the majority if not all computations carried out by ML, DL, and AI is directly connected to them. You can read my review paper of complex systems in medicine as it provides a general, fast, concise overview of the complexity as such.
Machine learning first.
You have to know the idea of machine learning, then learn several models including the basic neural network.After that, you can go to deep learning which mainly focuses on advanced neural networks.
Honestly, the best way to learn ML and DL is to start with mathematics. You must know very well algebra, vectors & matrices, have a decent knowledge of calculus, and most importantly know very well statistics and probability.
Do you know that the very first developed ML technique is well over one hundred years old? It is the linear regression that was developed within statistics.
Only when you have this knowledge, it is suitable to start with ML. I recommend you to use Python to write down programs as there are available very good libraries of ML techniques.
The last thing would be deep learning with its neuronal networks.
Lastly, I highly recommend you to study complex systems as the majority if not all computations carried out by ML, DL, and AI is directly connected to them. You can read my review paper of complex systems in medicine as it provides a general, fast, concise overview of the complexity as such.
Deep Learning may be viewed as stack-shaped machine learning.
Modern machine learning mainly relies on algebraic graph theory (Spectral Clustering/Spectral Regression), signal processing and advanced statistics which encompasses AI approaches (decision-trees, decision forests, etc.).
In a context of electrical power engineering, I suggest you to start with :
- Dörfler et al., "Electrical Networks and Algebraic Graph Theory: Models, Properties, and Applications ", 2017 - https://ece.uwaterloo.ca/~jwsimpso/papers/2017k.pdf
Regarding the links with Linux, have a look (for example) at NebulOS by Stickley et al.:
- Stickley et al., " NebulOS: A Big Data Framework for Astrophysics ", 2016 -
Article StratOS: A Big Data Framework for Scientific Computing
1. Linear algebra
2. Statistics
3. Data mining
4. Machine learning
5. Deep learning
1. Linear Algebra and Calculus
2. Statistic and Probability
3. Python (Good level)
4. R (Good level)
5. Neural Network
6. Fuzzy logic.
7. Complex System.
8. Machine learning.
9. Deep learning.
- Badges
- Science topic