Believe it or not, Coursera is probably the best place to learn about Machine learning and Deep learning online, and a big reason for that is Andrew Ng, who literally made Machine learning popular among developers. Well, Yes, and this course is part of their Advanced Machine Learning Specialization. But only last year October when the class relaunched, I decided to take it again, i.e watch all videos the second times, finish all homework and get passing grades for the course. Deep learning is inspired and modeled on how the human brain works. In fact, most of the sequence modelling problems on images and videos are still hard to solve without Recurrent Neural Networks. If you like this message, subscribe the Grand Janitor Blog's RSS feed. Just check out my own "Top 5-List". Getting Started with Neural Networks Kick start your journey in deep learning with Analytics Vidhya's Introduction to Neural Networks course! The old format only allows 3 trials in quiz, with tight deadlines, and you only have one chance to finish the course. What you'll learn Skip What you'll learn. [full paper ] [supporting online material (pdf) ] [Matlab code ] Papers on deep learning without much math. For models such as Hopfield net and RBM, it's quite doable if you know basic octave programming. However its become outdated due to the rapid advancements in deep learning over the past couple of years. For me, finishing Hinton's deep learning class, or Neural Networks and Machine Learning(NNML) is a long overdue task. And quite frankly I still don't grok some of the proofs in lecture 15 after going through the course because deep belief networks are difficult material. (Note: he was a physicist before working with neural networks. Like the course I just released on Hidden Markov Models, Recurrent Neural Networks are all about learning sequences – but whereas Markov Models are limited by the Markov assumption, Recurrent Neural Networks are not – and as a result, they are more expressive, and more powerful than anything we’ve seen on tasks that we haven’t made progress on in decades. Also, it spends a lot of time on some ideas (e.g. It is ideal for more complex neural networks like RNNs, CNNs, LSTMs, etc and neural networks you want to design for a specific purpose. You will work on case studi… Bestseller Created by Lazy Programmer Team, Lazy Programmer Inc. Learn how a neural network works and its different applications in the field of Computer Vision, Natural Language Processing and more. deep bayesian networks) which have largely fallen out of favor. This course will teach you almost everything you need to know as a Deep learning expert, not in the depth of the previous session but still good enough. Learning Deep Learning with Keras,a16z team’s reference links,Stanford’s CS 231n Convolutional Networks course website, and, of course, various Wikipedia pages concern-ingartificial neural networks. 10 Free Online course to learn Python in depth. Take at least Calculus I and II before you join, and know some basic equations from the Matrix Cookbook. Deep Learning on Coursera by Andrew Ng. PyTorch: Deep Learning and Artificial Intelligence - Neural Networks for Computer Vision, Time Series Forecasting, NLP, GANs, Reinforcement Learning, and More! This course provide the MOST in-depth look at neural network theory and how to code one with pure Python and Tensorflow. Only after you take that course, you should check these advanced courses to learn neural networks and deep learning in-depth. In this course, you will learn the foundations of Deep Learning, understand how to build neural networks, and learn how to lead successful machine learning projects. There is no doubt that Machine Learning is a tough subject, and in-depth knowledge, in particular, requires a lot of maths and complex terminology and very tough to master. Neural networks and deep learning are principles instead of a specific set of codes, and they allow you to process large amounts of unstructured data using unsupervised learning. I really like the way Kirill shows the intuitive part of the models, and Hadelin writes the code for some real-life projects. This is another impressive course from Coursera on Deep learning, didn’t I say that Coursera has the best Machine Learning course on the internet? If you don’t have 3 to 5 months to spare but want to learn deep learning in detail, then you should join this course. This course will demonstrate how neural networks can improve practice in various disciplines, with examples drawn primarily from financial engineering. I found myself thinking about Hinton's statement during many long promenades. Finally I made through all 20 assignments, even bought a certificate for bragging right; It's a refreshing, thought-provoking and satisfying experience. But I think understanding would come up at my 6th to 7th times going through the material. You bet! If you have any questions or feedback, then please drop a note. Together with Waikit Lau, I maintain the Deep Learning Facebook forum. You will also learn about Convolutional networks, RNNs, LSTM, Adam, Dropout, BatchNorm, Xavier/He initialization, and more. I mean, you are first introduced to the product, and then you deep dive into individual parts. A Verifiable Certificate of Completion is presented to all students who undertake this Neural networks course. Introduction to The Deep Learning A-Z™: Hands-On Artificial Neural Networks Course 504 - 507, 28 July 2006. Hello guys, if you want to learn Deep learning and neural networks and looking for best online course then you have come to the right place. I strongly recommend this course to anyone interested in Data Science and Deep Learning. Stories are compelling; they not just teach but also, inspire and you find them a lot in these excellent courses, which I am going to share with you about deep learning in-depth. Python vs. Java — Which Programming language Beginners should learn? Neural Networks and Deep Learning 2. In fact, Ng's Coursera class is designed to give you a taste of ML, and indeed, you should be able to wield many ML tools after the course. [1] To me, this makes a lot of sense for both the course's preparer and the students, because students can take more time to really go through the homework, and the course's preparer can monetize their class for infinite period of time. Templates included. 5786, pp. energy-based model and different ways to train RNN are some of the examples. They are seldom talked about these days. We will help you become good at Deep Learning. There are four reasons: All-in-all, Prof. Hinton's "Neural Network and Machine Learning" is a must-take class. Deep learning research also frequently use ideas from Bayesian networks such as explaining away. :) The downside: you shouldn't expect going through the class without spending 10-15 hours/week. 313. no. I was not so convinced by deep learning back then. In conclusion, this is an exciting training program filled with intuition tutorials, practical exercises, and real-World case studies. The homework requires you to derive backprop is still there. Even though Maths is an integral part of Deep Learning, I have chosen courses where you don’t need to learn complex Maths concepts, whenever something is required, the instructor explains in simple words. Plus, inside, you will find inspiration to explore new Deep Learning skills and applications. More than 16K Students have joined this course and you just need an Udemy account to enroll in this course. Confidently practice, discuss and understand Deep Learning concepts; How this course will help you? In these five courses, you will learn the foundations of Deep Learning, understand how to build neural networks, and learn how to lead successful machine learning projects. "Artificial intelligence is the new electricity." Which people these days still mix up with deep neural network (DNN). I firmly believe that every programmer should learn about Cloud Computing and Artificial Intelligence, as these two will drive the world in the coming years. In the first course, you'll learn about the foundations of neural networks, you'll learn about neural networks and deep learning. During the course you will also understand the applications of deep learning … Neural networks are a fundamental concept to understand for jobs in artificial intelligence (AI) and deep learning. Use This Guide To Sleep Smarter & Overcome Insomnia - Practical Tips, Including A Guided Meditation & Hypnosis (+ Ebook) Instructor: Kevin Kockot, M.A. The courses use Python and NumPy, a Python library for machine learning to build full-on non-linear. The goal of this course is to give learners a basic understanding of modern neural networks and their applications in computer vision and natural language understanding. Deep learning is a subset of Machine Learning which trains the model with huge datasets using multiple layers. We’ll emphasize both the basic algorithms … It’s by far the most comprehensive resource on deep learning. Deep Learning A-Z™ is structured around special coding blueprint approaches meaning that you won’t get bogged down in unnecessary programming or mathematical complexities and instead you will be applying Deep Learning techniques from very early on in the course. Then you would start to build up a better understanding of deep learning. If you have no basic background on either physics or Bayesian networks, you would feel quite confused. Many of my friends who have PhD cannot quite follow what Hinton said in the last half of the class. So one reason to take a class, is not to just teach you a concept, but to allow you to look at things from different perspective. I will chime in on the issue at the end of this review. I took the class last year October, when Coursera had changed most classes to the new format, which allows students to re-take. Here is the link to join this course online — Deep Learning A-Z™: Hands-On Artificial Neural Networks. You will build your knowledge from the ground up and you will see how with every tutorial you are getting more and more confident. It will also teach you how to install TensorFlow and use it for training your deep learning models. In this course you will be introduced to the world of deep learning and the concept of Artificial Neural Network and learn some basic concepts such as need and history of neural networks. The best part of this course I that it’s very well structured and moves step by step, which helps to build the complex deep learning and neural network concepts. So some videos I watched it 4-5 times before groking what Hinton said. Improving Deep Neural Networks: Hyperparameter tuning, Regularization, and Optimization 3. I admire people who could finish this class in the Coursera's old format. But more for second to third year graduate students, or even experienced practitioners who have plenty of time (but, who do?). If you only do Ng's neural network assignment, by now you would still wonder how it can be applied to other tasks. Companies using Tensorflow include Airbnb, Airbus, eBay, Intel, Uber and dozens more. e.g. If you are serious about deep learning, I strongly suggest you join this specialization and complete all five courses. LSTM would easily be your only thought on how to resolve exploding/vanishing gradients in RNN. Deep Learning A-Z™: Hands-On Artificial Neural Networks Course Catalog — The Tools — Tensorflow and Pytorch are the two most popular open-source libraries for Deep Learning. Here is the link to join this course — Introduction to Deep Learning. In this course, you will learn about how to create Deep Learning Algorithms in Python from two Machine Learning & Data Science experts: Kirill Eremenko and Hadelin de Pontes. Models such as Hopfield network (HopfieldNet), Boltzmann machine (BM) and restricted Boltzmann machine (RBM). If the subject matter is that tough, then how do you learn it better? A special mention here perhaps is Daphne Koller's Probabilistic Graphical Model, which found it equally challenging, and perhaps it will give you some insights on very deep topic such as Deep Belief Network. Students will gain an understanding of deep learning techniques, including how alternate data sources such as images and text can advance practice within finance. Check out his view in Lecture 10 about why physicists worked on neural network in early 80s. Inside Deep Learning A-Z™ you will master some of the most cutting-edge Deep Learning algorithms and techniques (some of which didn’t even exist a year ago), and through this course, you will gain an immense amount of valuable hands-on experience with real-world business challenges. It is deeper and tougher than other classes. This is Jeremy Howards’s classic course on deep learning. Science, Vol. We have also learned useful Python libraries like TensorFlow, Pandas, and Numpy, which can help you with data cleansing, parsing, and analyzing for your deep learning models. All of these make the class unsuitable for busy individuals (like me). You should realize performance number isn't everything. For more cool AI stuff, follow me at https://twitter.com/iamvriad. More about this course. Training Neural Network: Risk minimization, loss function, backpropagation, regularization, model selection, and optimization. The upside: you can still have all the fun of deep learning. Like the course I just released on Hidden Markov Models, Recurrent Neural Networks are all about learning sequences – but whereas Markov Models are limited by the Markov assumption, Recurrent Neural Networks are not – and as a result, they are more expressive and more powerful than anything we’ve seen on tasks that we haven’t made progress on in decades. Let me quantify the statement in next section. Python vs. JavaScript — Which is better to start with? It happens to many of my peers, to me, and sadly even to some of my mentors. Don't make the mistake! Learning Deep learning in-depth? About this course: Learn about artificial neural networks and how they’re being used for machine learning, as applied to speech and object recognition, image segmentation, modeling language and human motion, etc. In my case, I spent quite some time to Google and read through relevant literature, that power me through some of the quizzes, but I don't pretend I understand those topics because they can be deep and unintuitive. Even if you are used to the math of supervised learning method such as linear regression, logistic regression or even backprop, Math of RBM can still throw you off. - Andrew Ng, Stanford Adjunct Professor Deep Learning is one of the most highly sought after skills in AI. You’ve already written deep neural networks in Theano and TensorFlow, and you know how to run code using the GPU. Prof. Hinton's delivery is humorous. Hello guys, if you want to learn Deep learning and neural networks and looking for best online course then you have come to the right place. You will learn the basic building blocks of neural network and how it works layer by layer. That’s all about some of the best deep learning online courses to master neural networks and other deep learning concepts. For new-comers, it must be mesmerizing for them to understand topics such as energy-based models, which many people have hard time to follow. The Math is still not too difficult, mostly differentiation with chain rule, intuition on what Hessian is, and more importantly, vector differentiation - but if you never learn it - the class would be over your head. A Verifiable Certificate of Completion is presented to all students who undertake this Neural networks course. In my view, both Kapathy's and Socher's class are perhaps easier second class than Hinton's class. As you know, the class was first launched back in 2012. Like the course I just released on Hidden Markov Models, Recurrent Neural Networks are all about learning sequences – but whereas Markov Models are limited by the Markov assumption, Recurrent Neural Networks are not – and as a result, they are more expressive, and more powerful than anything we’ve seen on tasks that we haven’t made progress on in decades. Also check out my awesome employer: Voci. Not until 2 years later I decided to take Andrew Ng's class on ML, and finally I was able to loop through the Hinton's class once. You will practice ideas in Python and in TensorFlow, which you will learn on the course. It’s not the most advanced deep learning course out there, … While the previous one takes a bottom-up approach, this course takes a top-down approach. My Machine learning journey started a couple of years ago when I come to cross Andrew Ng’s excellent Machine Learning course on Coursera, It also happened to be Coursera’s first course as Andrew Ng is also one of the founders of Coursera. In Erweiterungen der Lernalgorithmen für Netzstrukturen mit sehr wenigen oder keinen Zwischenlagen, wie beim einlagigen Perzeptron, ermöglichen die Methoden des Deep Learnings auch bei zahlreichen Zwisc… For example, bias/variance is a trade-off for frequentist, but it's seen as "frequentist illusion" for Bayesian. Without wasting any more of your time, here is my list of best courses to learn Deep learning in-depth. Talking about social proof, this course has been trusted by more than 170,000 students, and it has, on average, 4.5 ratings from close to 23K ratings, which is just amazing. Further, RNNs are also considered to be the general form of deep learning architecture. He is another awesome instructor on the field of Deep Learning along with Andrew Ng of Coursera and Kirill Eremenko on Udemy. 10 Free Python Programming Books for Programmers, 9 Data Science and Machine Learning Courses for Beginners, Neuralink Is a Nightmare Dreamscape of a Medical Miracle, 5 Design Considerations For A Truly Conversational Chatbot, AI and Play, Part 1: How Games Have Driven Two Schools of AI Research, How The United States has Been Handing Its Lead in Artificial Intelligence to China. Simulated Consciousness, and Why I Believe It’s the Future of Interpersonal A.I. There is also a book with the same title which you can buy on Amazon. You will work on case studies from healthcare, autonomous driving, sign language reading, music generation, and natural language processing. Prof. Hinton teaches you the intuition of many of these machines, you will also have chance to implement them. Though, it’s expected that you have good knowledge of Python and Maths. Another suggestion for you: may be you can take the class again. This module introduces Deep Learning, Neural Networks, and their applications. That's said, you should realize your understanding of ML/DL is still .... rather shallow. Btw, if you are new to Machine learning then don’t start with these courses, the best starting point is still Andrew Ng’s original Machine Learning course on Coursera. May be you are thinking of "Oh, I have a bunch of data, let's throw them into Algorithm X!". P. S. — If you like to learn from free resources, then you can also check out this Deep Learning Prerequisites: The Numpy Stack in Python V2 free course on Udemy. If you finish this class, make sure you check out other fundamental class. The best part of the course is that you will hear from many top leaders in Deep Learning, who will share with you their personal stories and give you career advice, which is very inspiring and refreshing. The course explains the essentials of deep learning in a comprehensive way, before moving onto the more technical skills and exercises which will enable you to start building your very own neural networks. Coming back to Andrew’s Deep Learning Specialization, which is a collection of five courses focused on neural network and deep learning, as shown below: 1. Plus, inside you will find inspiration to explore new Deep Learning skills and applications. You can use any of these courses and online training to learn deep learning, but I highly recommend you to check Deep Learning specialization on Coursera by Andrew Ng and team. That's what I plan to do about half a year later - as I mentioned, I don't understand every single nuance in the class. Go for Hinton's class, feel perplexed by the Prof said, and iterate. Another story that inspired me a lot was of a Japanese farmer who used Google’s TensorFlow and Machine learning to filter and sort Cucumber on his farm, which apparently only his mother could do because of her years of experience. Introduction: Various paradigms of earning problems, Perspectives and Issues in deep learning framework, review of fundamental learning techniques. Some assignments made me takes long walks to think through. Sounds recursive? At this point, you already know a lot about neural networks and deep learning, including not just the basics like backpropagation, but how to improve it using modern techniques like momentum and adaptive learning rates. i.e. Deep Learning A-Z™: Hands-On Artificial Neural Networks online course has been taught by Kirill Eremenko and Hadelin de Ponteves on Udemy, this course is an excellent way to learn to create Deep Learning Algorithms in Python from two Machine Learning & Data Science experts. Neural Networks and Deep Learning. Try to grok. AI is not just for programmers but for everyone, and this is the best course to learn AI for all non-technical people like project managers, business analysts, operations, and event management team. The course starts with a recap of linear models and discussion of stochastic optimization methods that are crucial for training deep neural networks. Or what about deep belief network (DBN)? We are actually blessed that we have many excellent instructors like Andrew Ng, @Jeremey Howard’s, and Kirill Eremenko on Udemy around who are not just the expert of deep learning but also excellent instructors and teachers. I do recommend you to first take the Ng's class if you are absolute beginners, and perhaps some Calculus I or II, plus some Linear Algebra, Probability and Statistics, it would make the class more enjoyable (and perhaps doable) for you. This video that you're watching is part of this first course which last four weeks in total. Always seek for better understanding! As you read through my journey, this class is hard. So this piece is my review on the class, why you should take it and when. "Oh, we just want to use XGBoost, right! No wonder: at the time when Kapathay reviewed it in 2013, he noted that there was an influx of non-MLers were working on the course. Movies of the neural network generating and recognizing digits. Here is the link to buy his book — Deep Learning for Coders with fastai and PyTorch: AI Applications Without a PhD. If you don’t know, he is also one of the founders of Coursera, and his classic Machine learning course offered by Stamford is probably the first online course on Coursera. It cost around $399/year but its complete worth of your money as you get unlimited certificates. Of course, my mind changed at around 2013, but the class was archived. NNML is well-known to be much harder than Andrew Ng's Machine Learning as multiple reviews said (here, here). In that sense, NNML perfectly fit into the bucket. Convolutional Neural Networks 5. Here is the link to join this course — Data Science: Deep Learning in Python. Geoffrey Hinton’s course titled Neural Networks does focus on deep learning. And, as the number of industries seeking to leverage these approaches continues to grow, so do career opportunities for professionals with expertise in neural networks. ). I highly recommend this course to anyone who wants to know how Deep Learning really works. Recurrent Neural Networks (RNNs), a class of neural networks, are essential in processing sequences such as sensor measurements, daily stock prices, etc. Here is the link to join this course — Deep Learning Specialization. Of course, there are other ways: echo state network (ESN) and Hessian-free methods. But I still recommend NNML. It is, indeed. Course content. Data Science, Machine Learning, and Deep Learning are essential for understanding and using Artificial intelligence in many ways, and that’s why I am spending a lot of my spare time learning these technologies. More than the course, Andrew inspired me to learn about Machine Learning and Artificial intelligence, and ever since that, whenever I read him like on his Deep Learning course launch on Medium, I always get excited to learn more about this field. You will also find an in-depth explanation of maths behind ANN, which is very important for data scientists. You can also find me (Arthur) at twitter, LinkedIn, Plus, Clarity.fm. [1] It strips out some difficulty of the task, but it's more suitable for busy people. In this course, you will learn both! If you like this article, you may like my other Python, Data Science, and Machine learning articles as well: Thanks for reading this article so far. Apart from that classic course, Andrew has created a couple of more gems like AI For Everyone, which is again I recommend to every programmer and non-tech guys. The course is not just about boring theories; it’s very hands-on and interactive. Feedforward neural network: Artificial Neural Network, activation function, multi-layer neural network. (20170411) Fixed typos. Well, choose a course that can explain this complex topic in simple words. 1,164 students enrolled . It may take between 3 to 5 months, but it’s completely worth your time and more than 500K learners have already benefited from this specialization. Or is it still the best beginner class? One homework requires deriving the matrix form of backprop from scratch. Smooth up writings. Create Neural network models in Python using Keras and Tensorflow libraries and analyze their results. Learners these days are perhaps luckier, they have plenty of choices to learn deep topic such as deep learning. Talking about his course, it’s just the opposite of Andrew Ng’s Deep learning course. That doesn't mean you can go easy on the class : for the most part, you would need to review the lectures, work out the Math, draft pseudocode etc. Unlike Ng's and cs231n, NNML is not too easy for beginners without background in calculus. Same thing can be said about concepts such as backprop, gradient descent. But then he persisted, from his lectures, you would get a feeling of how/why he starts a certain line of research, and perhaps ultimately how you would research something yourself in the future. If you like these deep learning courses, then please share it with your friends and colleagues. In August 2016, Again, their formulation is quite different from your standard methods such as backprop and gradient-descent. I have chosen courses that are suitable for both beginners and developers with some experience in the field of Machine learning and Deep Learning. No wonder: many of these models have their physical origin such as Ising model. You will learn about Convolutional networks, RNNs, LSTM, Adam, Dropout, BatchNorm, Xavier/He initialization, and more. cs231n, cs224d and even Silver's class are great contenders to be the second class. It covers a lot of ground from basic to advanced deep learning concepts like ANN and CNN concepts. Earlier, I have shared the best data science course and today, I am going to share best deep learning online courses from Udemy, and Cousera. Video created by IBM for the course "Deep Learning and Reinforcement Learning". It's important to understand what's going on with your model. For me, finishing Hinton's deep learning class, or Neural Networks and Machine Learning(NNML) is a long overdue task. This is another awesome coursera specizliation to learn Deep learning. Check out my post "Learning Deep Learning - My Top 5 List", you would have plenty of ideas for what's next. Deep Learning Specialization by Andrew Ng and Team, Deep Learning A-Z™: Hands-On Artificial Neural Networks, Practical Deep Learning for Coders by fast.ai, Deep Learning for Coders with fastai and PyTorch: AI Applications Without a PhD, 5 Data Science and Machine Learning course in Python, 10 Resources to Learn Data Science in 2020, Top 5 Course to Learn Python for Beginners, Top 8 Python libraries for Data Science and Machine Learning, Top 5 Books to learn Python for Machine Learning. Deep Learning (frei übersetzt: tiefgehendes Lernen) bezeichnet eine Klasse von Optimierungsmethoden künstlicher neuronaler Netze (KNN), die zahlreiche Zwischenlagen (englisch hidden layers) zwischen Eingabeschicht und Ausgabeschicht haben und dadurch eine umfangreiche innere Struktur aufweisen. If you ever wanted a course that can teach you how to create your own neural network from scratch, then this is the course you should join. Hinton's perspective - Prof Hinton has been mostly on the losing side of ML during last 30 years. I also discuss one question which has been floating around forums from time to time: Given all these deep learning classes now, is the Hinton's class outdated? You easily make costly short-sighted and ill-informed decision when you lack of understanding. All of us, beginners and experts include, will be benefited from the professor's perspective, breadth of the subject. Another reason why the class is difficult is that last half of the class was all based on so-called energy-based models. But learning them give you breadth, and make you think if the status quote is the right thing to do. Hinton, G. E. and Salakhutdinov, R. R. (2006) Reducing the dimensionality of data with neural networks. Another more technical note: if you want to learn deep unsupervised learning, I think this should be the first course as well. Inside Deep Learning A-Z™ you will master some of the most cutting-edge Deep Learning algorithms and techniques (some of which didn't even exist a year ago) and through this course you will gain an immense amount of valuable hands-on experience with real-world business challenges. MOOCs In April 2017, David Venturi collected an im-pressivelist of Deep Learning online courses along with ratings data. Once you think about them, they are tough concepts. It always give you the best results!" And each of the five courses in the specialization will be about two to four weeks, with most of them actually shorter than four weeks. If you are not comfortable with Python yet, I suggest you take one of the top Python courses I have suggested before. As you know, the class was first launched back in 2012. not so convinced by deep learning back then, Review of Ng's deeplearning.ai Course 4:…, Review of Ng's deeplearning.ai Course 3:…, Review of Ng's deeplearning.ai Course 2:…. Which programming language works best with PyTorch? As I explained before, NNML is tough, not exactly mathematically (Socher's, Silver's Maths are also non-trivial), but conceptually. Feedforward neural networks are the simplest versions and have a single input layer and a single output layer. PyTorch is an excellent framework for getting into actual machine learning and neural network building. This course, you will get you started in building your first artificial neural network using deep learning techniques. And, if you find Coursera courses, specialization, and certifications useful then I suggest you join the Coursera Plus, a great subscription plan from Coursera which gives you unlimited access to their most popular courses, specialization, professional certificate, and guided projects. Sequence Models Andrew follows a bottom-up approach, which means you will start from the smallest component and move towards building the product. In fact, in the course, we will be building a neural network from scratch using PyTorch. Many concepts in ML/DL can be seen in different ways. Structuring Machine Learning Projects 4. Suppose you just want to use some of the fancier tools in ML/DL, I guess you can just go through Andrew Ng's class, test out bunches of implementations, then claim yourself an expert - That's what many people do these days. If you learn RNN these days, probably from Socher's cs224d or by reading Mikolov's thesis. No?