12 answers
Mohamed
Student
Dar es Salaam, Tanzania
1
Question
Updated
1735 views
What is better to start among the following, machine learning, deep learning or AI? What are the requerements for a beginner?
Life sciences student (applied DNA technology), python beginner #science #technology #python #coding #AI #ML #neuralnetworks #programming
Login to comment
12 answers
Ramanandan NK
Software engineer
4
Answers
Bengaluru, Karnataka, India
Updated
Ramanandan’s Answer
As a beginner, it's crucial to start with the basics and gradually progress. Here's a concise overview of the three main areas:
1. Machine Learning (ML): ML is a part of artificial intelligence (AI) that enables systems to learn and enhance from experience without specific programming. It concentrates on creating algorithms that can adapt and change when introduced to new data.
2. Deep Learning (DL): DL is a machine learning branch that uses artificial neural networks, inspired by the human brain, to learn from vast amounts of data. It can manage and process more intricate data types like images, sound, and text.
3. Artificial Intelligence (AI): AI is a wider concept that involves machines or software imitating human intelligence. AI can be rule-based and doesn't always learn from data, unlike ML and DL.
Considering these definitions, it's logical to start with Machine Learning. ML offers the essential knowledge required to comprehend more advanced concepts in Deep Learning and other AI subjects.
To begin your Machine Learning journey, follow these steps and requirements:
1. Mathematics: A basic understanding of linear algebra, calculus, probability, and statistics helps in grasping the principles of machine learning algorithms.
2. Programming: Get acquainted with a programming language, preferably Python, as it's popular in the data science and machine learning community due to its simplicity and the presence of machine learning libraries like scikit-learn, TensorFlow, and PyTorch.
3. Basic knowledge of data structures and algorithms: This assists in writing efficient code and comprehending the computational complexity of various machine learning algorithms.
4. Start with the basics: Learn about different machine learning algorithms such as supervised learning, unsupervised learning, and reinforcement learning. Begin with straightforward algorithms like linear regression or decision trees before advancing to more complex ones.
5. Practical Implementation: While theory is vital, practical implementation is equally important. Engage in simple projects or Kaggle competitions to understand data pre-processing, model selection, model training, and performance evaluation.
6. Keep Learning: Machine learning is an ever-changing field. Stay updated on new algorithms, techniques, and best practices.
Patience and persistence are key in this complex field. With consistent effort and time, you'll make progress.
1. Machine Learning (ML): ML is a part of artificial intelligence (AI) that enables systems to learn and enhance from experience without specific programming. It concentrates on creating algorithms that can adapt and change when introduced to new data.
2. Deep Learning (DL): DL is a machine learning branch that uses artificial neural networks, inspired by the human brain, to learn from vast amounts of data. It can manage and process more intricate data types like images, sound, and text.
3. Artificial Intelligence (AI): AI is a wider concept that involves machines or software imitating human intelligence. AI can be rule-based and doesn't always learn from data, unlike ML and DL.
Considering these definitions, it's logical to start with Machine Learning. ML offers the essential knowledge required to comprehend more advanced concepts in Deep Learning and other AI subjects.
To begin your Machine Learning journey, follow these steps and requirements:
1. Mathematics: A basic understanding of linear algebra, calculus, probability, and statistics helps in grasping the principles of machine learning algorithms.
2. Programming: Get acquainted with a programming language, preferably Python, as it's popular in the data science and machine learning community due to its simplicity and the presence of machine learning libraries like scikit-learn, TensorFlow, and PyTorch.
3. Basic knowledge of data structures and algorithms: This assists in writing efficient code and comprehending the computational complexity of various machine learning algorithms.
4. Start with the basics: Learn about different machine learning algorithms such as supervised learning, unsupervised learning, and reinforcement learning. Begin with straightforward algorithms like linear regression or decision trees before advancing to more complex ones.
5. Practical Implementation: While theory is vital, practical implementation is equally important. Engage in simple projects or Kaggle competitions to understand data pre-processing, model selection, model training, and performance evaluation.
6. Keep Learning: Machine learning is an ever-changing field. Stay updated on new algorithms, techniques, and best practices.
Patience and persistence are key in this complex field. With consistent effort and time, you'll make progress.
Rajesh Bhatawadekar
10
Answers
India
Updated
Rajesh’s Answer
My advise would be to pick up a generic language, like Python, get very familiar with it, also have a good understanding of mathematics and statistics. All the domains you mentioned will require these basic knowledge, and then you can pick lot of libraries out there to explore the fields. Lot of online courses out there that cover these topics from beginner, intermediate, to advanced level. You also have an option to complete a real project.
Learn Python
Take online course/s
Complete a project using sample data (lot of options available online, including capstone projects
Identify a problem that you want to solve with the technology and go for it !
Rajesh recommends the following next steps:
Anil Varma Kucherlapati
Data Scientist
1
Answer
Bengaluru, Karnataka, India
Updated
Anil Varma’s Answer
Machine Learning refers to the set of algorithms which can learn and infer patterns in the data provided. Deep learning is a type of implementation of Neural networks, which is a ML algorithm. While Deep learning is popular and applicable for many usecases there are many other ML algorithms which might perform better, are more explainable and can be easily trained depending on the type of problem. Starting with ML will give you a good understanding & appreciation of the field and the flexibility to use & specialize in deep learning if really interested. You can also expand to AI which is focused on mimicking human behavior and leverages ML.
Knowledge of Math and Statistics concepts will be very helpful in building a strong foundation and to get a good understanding of the internal mechanism of these algorithms. Programming languages like Python and R are easy to pick up and will let you implement them
Knowledge of Math and Statistics concepts will be very helpful in building a strong foundation and to get a good understanding of the internal mechanism of these algorithms. Programming languages like Python and R are easy to pick up and will let you implement them
Thanks for clear explaination🤗
Mohamed
adarsh raj
KARNATAKA
3
Answers
Bengaluru, Karnataka, India
Updated
adarsh’s Answer
Hi,
It's the question, I believe all the folks who want to start learning machine learning concepts, AI concepts want to know.
As a beginner, it's always important to understand your motive to learn a skill.
Python is the most widely used language nowadays and it has been promoted as one of the most desired language skills to have.
It's been ranked as one of the top language skills to have in the current IT world and is a good start for the coding.
To start with for beginners, learn python and try to practice coding from it. You can also take up few python courses online which are short and crisp and also help you learn faster.
R is also a very remarkable language and is mostly used for statistical analysis tools and problems. R surely would help to analyze and predict solutions to the problems fed to the code.
To start with for Machine leaning, it's best to get to learn python and learn all the basic concepts, w.r.t machine learning & its algorithms. It's important to also understand the difference between Machine Learning, Artificial Intelligence, Deep Learning & Data Science.
Learning the concepts and understanding which algorithm to use in which problem statement is very important. This will only be possible when you go through the various algorithms of Machine learning, its types and ways of implementation.
Once you learn all concepts and know-how to start implementing, them, it's important to again practise more and more.
Reference link: https://www.youtube.com/watch?v=9f-GarcDY58
It's the question, I believe all the folks who want to start learning machine learning concepts, AI concepts want to know.
As a beginner, it's always important to understand your motive to learn a skill.
Python is the most widely used language nowadays and it has been promoted as one of the most desired language skills to have.
It's been ranked as one of the top language skills to have in the current IT world and is a good start for the coding.
To start with for beginners, learn python and try to practice coding from it. You can also take up few python courses online which are short and crisp and also help you learn faster.
R is also a very remarkable language and is mostly used for statistical analysis tools and problems. R surely would help to analyze and predict solutions to the problems fed to the code.
To start with for Machine leaning, it's best to get to learn python and learn all the basic concepts, w.r.t machine learning & its algorithms. It's important to also understand the difference between Machine Learning, Artificial Intelligence, Deep Learning & Data Science.
Learning the concepts and understanding which algorithm to use in which problem statement is very important. This will only be possible when you go through the various algorithms of Machine learning, its types and ways of implementation.
Once you learn all concepts and know-how to start implementing, them, it's important to again practise more and more.
adarsh recommends the following next steps:
Ooooh thank u 4 clear explaination for my beloved career
Mohamed
Dylan Shiramizu
4
Answers
Updated
Dylan’s Answer
The previous answers have really good resources listed as well as steps to get started in AI/Machine Learning/ Deep Learning! This may be a little late for you seeing that it has been a couple years since you posted but for those that stumble upon this thread and are new and want to dip your toes into machine learning, here are some more ways to get started.
Like the others said, I would advise that you have strong foundation in python.
Learning the math behind all of the Machine Learning algorithms will be beneficial.
Then its just a matter of choosing which Machine Learning techniques you would like to learn. Each technique has different use cases. Knowing what technique you need for what you would like to apply it to would be beneficial to find out where to start.
https://www.udacity.com/
Udacity has many courses on AI, Machine Learning, and Deep Learning. The courses are very helpful and explain the math behind the algorithms as well as the applications of each technique. Udacity also provides introductory courses for both pytorch and tensorflow!
https://www.coursera.org/
Coursera also has a plethora of AI/Machine Learning/Deep Learning courses available. For deep learning, I would recommend the courses from DeepLearning.ai by Andrew Ng. (https://www.coursera.org/specializations/deep-learning)
Coursera courses can be expensive at times but you receive a 2 week free trial to every course... so do what you will with that information.
Like the others said, I would advise that you have strong foundation in python.
Learning the math behind all of the Machine Learning algorithms will be beneficial.
Then its just a matter of choosing which Machine Learning techniques you would like to learn. Each technique has different use cases. Knowing what technique you need for what you would like to apply it to would be beneficial to find out where to start.
https://www.udacity.com/
Udacity has many courses on AI, Machine Learning, and Deep Learning. The courses are very helpful and explain the math behind the algorithms as well as the applications of each technique. Udacity also provides introductory courses for both pytorch and tensorflow!
https://www.coursera.org/
Coursera also has a plethora of AI/Machine Learning/Deep Learning courses available. For deep learning, I would recommend the courses from DeepLearning.ai by Andrew Ng. (https://www.coursera.org/specializations/deep-learning)
Coursera courses can be expensive at times but you receive a 2 week free trial to every course... so do what you will with that information.
Jack Pettrakool
Network Development Principal Engineer at Dell
3
Answers
Santa Clara, California
Updated
Jack’s Answer
AI is just general term while machine learning, deep learning, and whatever learning are overall method/algorithm to implement AI. You can start from any of them, but at the end you need to consider of your goal, amount of data, time for training, and purpose of AI. You should pick one that fit your need the best.
But if you just want to learn about it. There are several on-line courses, but I recommended https://course.fast.ai/ since it walks you through basic and give you idea (how-to), including example to use the libs.
It may require you to know about computer programming language like: python, java, etc; and mathematics like matrices, vector, tensor, among other things.
Thanks much!⭐⭐⭐⭐⭐
Mohamed
Kunwar Vikrant
Software Engineer
5
Answers
Bengaluru, Karnataka, India
Updated
Kunwar’s Answer
AI is actually a superset of Machine Learning which in turn is a superset of deep learning. If you want to get into this field, you need to have knowledge of some programming language. The most used programming language today in this field is python. Once you are confident in the language, you can start learning about the subjects.If you are in college, i'll suggest you take up the courses in the semester if they are offered. If not there are a lot of quality content available on the websites like coursera, udemy etc. You can have a look through them. If you really want to get into the data science field and learn ML and/or DL then i suggest you have a look at the course offered a team at the following link,
https://www.appliedaicourse.com/
https://www.appliedaicourse.com/
Thanks alot😊
Mohamed
Vibush Shanmugam
Engineer
3
Answers
Bengaluru, Karnataka, India
Updated
Vibush’s Answer
There are a number of courses available online for machine learning and deep learning. Take some time to reach about machine learning and understand what it is before you choose to take it up. There are great courses available on sites like coursera and udemy.
A great course to begin with is https://www.coursera.org/learn/machine-learning . This course covers all the basics you need to understand machine learning as well as gives you some hands on exercises to help assess your knowledge.
A great course to begin with is https://www.coursera.org/learn/machine-learning . This course covers all the basics you need to understand machine learning as well as gives you some hands on exercises to help assess your knowledge.
Thanks alot engineer😊
Mohamed
Hariprasad Ashok
bussiness analyst-intern
5
Answers
Bengaluru, Karnataka, India
Updated
Hariprasad’s Answer
Machine learning is basically a way to make a computer learn from it's experience just like Humans in this case data is the experience.So there are different methodologies and models .
I feel you must improve python skills and working with data skills primarily.
Later you should take up some machine learning courses on Udemy and Coursera.
https://www.coursera.org/learn/machine-learning .
Work on hands on projects for better understanding.
I feel you must improve python skills and working with data skills primarily.
Later you should take up some machine learning courses on Udemy and Coursera.
https://www.coursera.org/learn/machine-learning .
Work on hands on projects for better understanding.
Nievedha
Student
Melbourne, Victoria, Australia
5
Questions
Updated
Nievedha’s Answer
As a beginner in AI, ML and DL, I faced the same difficulties. I started by learning the difference between the terms. Basics such as the history and how each component works. This made me understand the path that I am interested in.
It is good to have an understanding of programming languages such as Python. Math is also needed for the understanding of the algorithms and how they can be applied in various situations.
I have just started a free course in Artificial Intelligence and Machine Learning at Amity Future Academy(https://learningweek.amityfutureacademy.com/login/index.php). There are also alternate platforms such as Coursera, Pluralsight and Udemy that offers amazing courses. Hope this helps!
It is good to have an understanding of programming languages such as Python. Math is also needed for the understanding of the algorithms and how they can be applied in various situations.
I have just started a free course in Artificial Intelligence and Machine Learning at Amity Future Academy(https://learningweek.amityfutureacademy.com/login/index.php). There are also alternate platforms such as Coursera, Pluralsight and Udemy that offers amazing courses. Hope this helps!
Thanks alot🙂
Mohamed
Vidhi Agrawal
Network Engineer
3
Answers
Bengaluru, Karnataka, India
Updated
Vidhi’s Answer
To start with machine learning, deep learning or AI just start by learning the difference between all those that you have mentioned.
The difference between all this will help you in various ways, I have seen people getting confused between them.
Then start learning about the algorithms for this your mathematics should be good.
Once you are done with this identify which among those is of your best interest.
Then,I would say Udacity is the best platform to get your basics clear.
The difference between all this will help you in various ways, I have seen people getting confused between them.
Then start learning about the algorithms for this your mathematics should be good.
Once you are done with this identify which among those is of your best interest.
Then,I would say Udacity is the best platform to get your basics clear.
Thanks for your contribution😊
Mohamed
Vivek T V
Software Development
8
Answers
Bengaluru, Karnataka, India
Updated
Vivek’s Answer
Begin with AI fundamentals: Grasp the idea and its scope. AI (Artificial Intelligence) is an extensive term that signifies the ability of machines to imitate human intellect. This knowledge will serve as a strong base for comprehending subsequent subjects.
Proceed to Machine Learning: Machine Learning (ML) is a part of AI. It involves instructing computers to learn from information to make choices or forecasts. For the majority of individuals, machine learning is an excellent starting point since it offers essential tools, algorithms, and insights into how AI functions. Explore various ML models (Supervised, Unsupervised, Reinforcement Learning), algorithms (Linear Regression, Logistic Regression, SVM, Decision Trees, Random Forests, etc.), and practice implementing them on different datasets.
Delve into Deep Learning: Deep Learning (DL) is a subset of Machine Learning, comprising a neural network with at least three layers. These networks try to replicate human brain behavior—though far from equaling its capacity—to learn from vast data quantities. While a single-layer neural network can make rough predictions, extra hidden layers can enhance accuracy. Deep Learning powers numerous AI applications and services that boost automation, completing tasks without human input.
Furthermore, regardless of your chosen path, you should become acquainted with the following:
Programming: Python is the most prevalent language in this field, but R and Java are also widely utilized.
Mathematics and Statistics: A basic understanding of linear algebra, calculus, and statistical methods is beneficial, as they form the foundation of AI/ML algorithms.
Data handling and manipulation: Get to know tools like Pandas, Numpy, and data formats such as CSV, XLSX, JSON, and XML.
Data Visualization: Employ tools like Matplotlib and Seaborn to better visualize your data and outcomes.
Tools and libraries: Familiarize yourself with Scikit-learn, TensorFlow, PyTorch, Keras, and other useful libraries.
Proceed to Machine Learning: Machine Learning (ML) is a part of AI. It involves instructing computers to learn from information to make choices or forecasts. For the majority of individuals, machine learning is an excellent starting point since it offers essential tools, algorithms, and insights into how AI functions. Explore various ML models (Supervised, Unsupervised, Reinforcement Learning), algorithms (Linear Regression, Logistic Regression, SVM, Decision Trees, Random Forests, etc.), and practice implementing them on different datasets.
Delve into Deep Learning: Deep Learning (DL) is a subset of Machine Learning, comprising a neural network with at least three layers. These networks try to replicate human brain behavior—though far from equaling its capacity—to learn from vast data quantities. While a single-layer neural network can make rough predictions, extra hidden layers can enhance accuracy. Deep Learning powers numerous AI applications and services that boost automation, completing tasks without human input.
Furthermore, regardless of your chosen path, you should become acquainted with the following:
Programming: Python is the most prevalent language in this field, but R and Java are also widely utilized.
Mathematics and Statistics: A basic understanding of linear algebra, calculus, and statistical methods is beneficial, as they form the foundation of AI/ML algorithms.
Data handling and manipulation: Get to know tools like Pandas, Numpy, and data formats such as CSV, XLSX, JSON, and XML.
Data Visualization: Employ tools like Matplotlib and Seaborn to better visualize your data and outcomes.
Tools and libraries: Familiarize yourself with Scikit-learn, TensorFlow, PyTorch, Keras, and other useful libraries.