 Machine learning, deep learning, and AI have become the new buzzwords of the decade, and everybody's hopping on that bandwagon. The number of papers published every month on Archive has increased over 8 folds since 2014. That's quite a bit, and it makes you wonder as a reader, how on earth can we keep up with research? On my AI channel, I've made over 60 videos, many of them tackling trending deep learning research, and in this video, I'll share my sources on keeping up with breaking machine learning and deep learning research, so you can always be in the loop. This is Code Emporium, so let's get started. I need to start out with a good old archive. It's a repository of hundreds of thousands of papers in physics, mathematics, computer sciences, and many other fields. And it should be your go-to for AI research. But look at this. It's a mess. It's really difficult to know what's important and what's not. So it's best to go to Archive Sanity, which is basically archive but cleaner. You can view the most recent papers. You can view the top papers. These are papers with a large number of citations. In other words, these papers have been used in other research in the respective field. And you can see the papers that people are talking about. So what's the big hype? This section has mostly new papers that introduce existing concepts and could eventually lead to another branch of research in the field. Basically Archive Sanity sheds a lot of time when you're trying to determine what paper should I read next. Oh, and fun fact. This repository was created by Andrzej Karpathy, because he was bored. Looks like the position of director at AI at Tesla can become very monotonous. Oh, and he has a YouTube channel on SweetCubing that I used to follow. That's pretty neat. So all you SweetCubers out there, check out Badminfesto. It's worth your while. So Archive Sanity, it's my go-to place to see what's popping. My go-to now for big ideas is Twitter. It's important to follow people who tweet mostly AI. I say this because following everyone that follows you can just decrease the quality of your feed. I'd recommend starting out by following big people in AI. Like Coursera founder Andrew Ong, computer vision pioneer Fei-Fei Li, the director of AI at Tesla, Andrzej Karpathy, Yang Likun is great to follow, he's the man behind convolutional neural networks, and you can also follow Ian Goodfellow, the man behind generative adversarial networks. In addition to these guys, you can follow anyone interesting that you come across. They could be tweeting about some projects they're working on, or some ideas they just want to share. It could be research work they just dropped, or some are just bored. In fact a recent video I made on the evolution of face generation research was based off a tweet by Ian Goodfellow. He conveniently shared links to five landmark papers in modeling something called generative adversarial networks, which is the technology used to generate those fake faces, or faces of people who don't exist. If you want to hear me explain these five papers in 12 minutes, and you want to know more about generative models like generative adversarial networks, then check out the links in the description to my videos. Anyways, I suggest creating a separate Twitter for just your AI following so that every tweet in your feed will be AI related. So at this point you have Archive Sanity as your go-to for breaking AI research, and Twitter as your go-to for big ideas in the field. Moving on, YouTube. And yes, YouTube is considered social media for the most part. Let's name a few channels that you should follow. Two minute papers. As in the name, this channel discusses new and trending research papers in about 5 minutes. He explains these papers in a very easy to understand way that leaves you curious to learn more. He states the abstract, highlights the experiment, and shows us results with really nice visuals. No matter what your level of machine learning understanding is, I definitely sub to two minute papers to know the most groundbreaking research when it drops. Next up we have Three Blue One Brown. This is traditionally a math channel. Grant Sanderson, the main man behind the scenes, teaches mathematical concepts using stunning visuals. For all you people aspiring to do something in the field of AI, I highly recommend checking out his playlist titled The Essence of Calculus. He explains concepts like limits and determinants in such a visual way that you'll look at them in a completely different light. And these concepts are fundamental to many machine learning algorithms we see today. So sub to Three Blue One Brown for establishing your math foundations. You may think you know something, but when you see his videos, you'll probably think again. Siraj Ravall. Of all the content creators I've mentioned, this one is the most similar to my own channel, in the sense that we both make content on AI. Siraj has been on YouTube making content on AI for about three years at the time of making this video, and he's the biggest on this list. Wizard of the Week. That's what he calls his followers. Regardless of what many say about him, I personally like his content. He has these weekly coding challenges that are pretty fun, and it gets you into the programming aspect of AI. Nowadays, he targets more of a beginner intermediate level audience, so I use his channel as a means for finding out breaking news in the field. I watch his videos as an introduction to the concept, and then read papers about it if I'm more curious. For some cool programming projects and computer vision, I would go check out Sentex. He's been a creator for years now, and he's the go-to guy for Python programming on YouTube. But he's recently made a number of hands-on programming tutorials and machine learning. You can check out his very cool playlist on self-driving cars that I thought was very impressive. He'll walk you right through everything from the very basics, so check him out. If I were to put these guys on the line where the right is purely theory and the left side is purely application-oriented, then I'd put three blue one brown and two-minute papers on the theory side, Sentex on the programming side, and Siraj towards the middle slash programming side. He's got some theoretical videos too, though. And there's also me. For now, I make quite theoretical videos, explaining algorithms, concepts, and research papers in detail. I do have some programming videos, but not that much as I'm making this video. But who knows how I'll evolve. Yes, I'm aware I sound like a Pokemon. So until now, ArchiveSanity used for breaking AI research, Twitter for getting big ideas, and YouTube for both AI research and tutorials on getting yourself up to speed on any concept. Like YouTube, another type of social media you can use for expanding your AI knowledge is Quora. It's basically Yahoo! Answer is done right. Now, I don't usually use this as a source of groundbreaking events in the fields of artificial intelligence, but there are surprisingly some really high-qualified AI scientists on the platform that know how to explain certain concepts. Let's mention a few. Zishan Zia, probably my favorite writer on the platform, who isn't me, of course. He consistently writes high-quality detailed answers on machine learning, deep learning, computer vision, and other branches of artificial intelligence. He has a PhD in computer vision, so all you vision folks out there strongly recommend you follow him. Next up, we have Chris Lurs. He's my number two, and is probably the most underrated data scientist on the platform. I've been reading his answers and also Zishan's answers through the last year, and I've learned so much doing so. He has amazing answers on statistical concepts and very high-quality content in general. Give him a follow. His content speaks for himself. Prasoon Goyle. He answers more technical questions in machine learning and general artificial intelligence. He's a frequent writer, has great content, a PhD student at UT Austin, and a lot of my followers also follow him. Next up, Ajith Rajshakran. At times, very succinct, but at other times, very descriptive. He's answered quite a number of questions on machine learning and natural language processing. If you're interested in natural language processing, then I strongly recommend you follow him. Shreither Madhavan. With a tad over 100 answers on the platform at the time of making this video, he has not answered much on the platform, at least compared to the other people I've mentioned. However, the quality and details of his answers to questions on machine learning is amazing. Reading his answers, you can tell that he has experience. Last but not least, Ajay Hathor. Yep, me. I write about machine learning, deep learning, data sciences, and artificial intelligence. Although I'm not as active now. Whoops. There are many more people out there, and this is just my personal list. I value quality over quantity, and I didn't want to just list out hundreds of names. I recommend you follow these people, look at who they are, and who they follow, and augment your list from there. While we're on the topic of social media, I'll also mention Facebook groups. You may have already deleted your Facebook, but there are some groups out there that are very helpful. Just note that some of these groups may require you to answer a few questions before joining, like, why do you want to join, or what will you contribute after joining? As long as you're chill, you should be able to get into any of these groups. If you don't see them all now, it's fine. I'll link everything in the description. Another great place to look for groundbreaking content is where it all starts, tech blogs. Technically, they are sources of information of all the hype on social media, but there are so many blogs out there, and each blog pumps out a ton of information, so much so that it can be difficult to know what's important in the field and what isn't. If you're a tech nerd and want to know about technology advancements, big or small, then add tech blogs to your source of research information. And that's it. That is my non-exhaustive list of resources. To recap, use Archive Sanity for getting to know breaking AI research in the form of research papers. Use Twitter for getting big ideas for personal projects. Use YouTube for both understanding groundbreaking AI research and for tutorials on any concept. It could be something as simple as what are support vector machines, for example. Use Quora for asking the big questions and understanding concepts and breaking research in machine learning, statistics, and data sciences. You can be a part of a number of Facebook groups that specialize in artificial intelligence and deep learning. So you can participate in forums and ask and answer questions in the field. You can actually do something similar with Reddit, too. And then there are tech blogs for the nerdy folks out there to keep up with research and technology, and also how industries are making use of breaking AI research. And that's the list. Now you know how to keep up with trends in the field of AI. Remember, the number one way to keep up is to subscribe to Code Emporium and see his dojo. Because we do everything you'll ever need to know, ever. The links to everything I've mentioned is down in the description below the video. Thank you so much for watching until the end, and I'll see you in the next one. Bye-bye.