I’ve had friends ask me how they can get started in Machine Learning, Deep Learning, NLP etc. and this is the answer I usually give them- read more. Read widely and read regularly. And don’t stress so much about the results. As Andrew Ng says-
… the most important thing is to keep on learning. Not just for a few months, but for years.
Every Saturday, you will have a choice between staying at home and reading research papers/implementing algorithms, vs. watching TV. If you spend all Saturday working, there probably won’t be any short-term reward, and your current boss won’t even know or say “nice work.” Also, after that Saturday of hard work, you’re not actually that much better at machine learning. But here’s the secret: If you do this not just for one weekend, but instead study consistently for a year, then you will become very good.
I’ve always found that following newsletters and feeds help to keep me in the loop with respect to books, new papers, industry news, tech, online classes, etc. And they often help me find other things to read, which then help me find other…
Without further ado, here’s my list of preferred ML/AI feeds:
- The Wild Week in AI- by wunderkind Denny Britz, good with tech and doesn’t skimp out on the math
- Import AI- good meta-analyses of recent DL/RL papers and I especially enjoy the scoop about the business of AI from someone in the Valley
The business of AI:
- a16z- Andreessen Horowitz’s VC firm has a newsletter that covers AI and other tech happenings
- Nathan Benaich does huge posts covering quarterly AI trends in technology, research and the venture capital market
- Benedict Evans
I’ve never been big on social media, but Twitter is great as a news aggregator. Start following researchers, data scientists, etc that you like. There’s always something interesting going on there. Most of them have blogs which you should read too. Here’s a few-
- @Sebastian Ruder - Transfer Learning and NLP
- @Francois Chollet - Author of Keras. Opinionated and entertaining
- @Denny Britz - Google Brain resident. Stay in the loop with new DL stuff
- @hardmaru - RL and Evolutionary approaches
- @jeremyphoward - Practical deep Learning
- @Ferenc Huszar - ML theory, Bayesian
- @Miles_Brundage - AI policy and safety
- @Hacker News - The tech water cooler
- @Data Tau - The data science water cooler
Or you could just follow my feed and stay on top of everything #ShamelessPlug
- PyData - regular uploads of PyData conference talks. Diverse set of topics, definitely worth a look
- The Talking Machines- Very technical and great way to expand your ML horizon. Just remember that there’s only so much math you’re going to remember from a podcast. Niel Lawrence and Katherine Gorman bring on interesting personalities from the world of machine learning and keep the conversations engaging
Don’t be surprised if there’s a lot of overlap in the content, there’s only so much that happens in a week. But mostly, don’t stress about not being exhaustive in covering everything, you’ll get there.
Of course, this is if you just want to know more in general. As one of my professors says, it always helps to have a motivating question when learning something. Then you’ll figure out what you need to do to fill in the blanks. If you’re looking for more targeted information, you might want to check out my list of recommended MOOCs
So that’s a wrap. I’ll keep adding more links to this list as and when I find new content. Anyway, let me know in the comments if there are any great newsletters/feeds that help you stay in the know.
Update: Sandeep Kandukuri on LinkedIn suggested I add the Machine Learning subreddit. I have to agree that it’s a great source of information that I don’t find anywhere else. Especially their AMAs like this and this.