Technology, A.I., Careers
There is a subfield of robotics research called “sim-to-real” (sim2real) whereby one attempts to solve a robotic task in simulation, and then get a real robot to do the same thing in the real world. My team at Google utilizes Sim2Real techniques extensively in pretty much every domain we study, including locomotion and navigation and manipulation . The arguments for doing robotic research in…
One of my ML research mentees is following OpenAI's Spinning up in RL tutorials (thanks to the nice folks who put that guide together!). She emailed me some good questions about the basics of Reinforcement Learning , and I wanted to share some of my replies on my blog in case it helps further other student's understanding of RL. The classic Sutton and Barto diagram of RL Your “ How to Understand…
Eric Jang, "Ten Apes", Jun 19 2021. NFT "drop" coming soon. Andy Warhol once said, “Art is what you can get away with.” I interpret the quote as a nihilistic take on “beauty is in the eye of the beholder” — a urinal you found in the junkyard can be considered art, so long as you convince someone to buy it, or showcase it in a museum. All that matters is what other people see in it and what buyers…
This essay is about how the powerful want to become countries, and the implications of cryptocurrencies on the sovereignty of nations. I’m not an economics expert: please leave a comment if I have made any errors. Money allows goods, services, and everything else under the sun to be assigned a value using the same unit of measurement. Without money, society reverts to bartering , which is highly…
This is the text version of a talk I gave on March 12, 2021, at the Brown University Robotics Symposium. As always, all views are my own, and do not represent those of my employer. I'm going to talk about why I believe end-to-end Machine Learning is the right approach for solving robotics problems, and invite the audience to think about a couple interesting open problems that I don't know how to…
“Traducción a Español” Biologically Plausible Deep Learning (BPDL) is an active research field at the intersection of Neuroscience and Machine Learning, studying how we can train deep neural networks with a "learning rule" that could conceivably be implemented in the brain. The line of reasoning that typically motivates BPDL is as follows: A Deep Neural Network (DNN) can learn to perform…
My ML mentees often ask me some variant of the question "how do you choose which papers to read from the deluge of publications flooding Arxiv every day?” The nice thing about reading most ML papers is that you can cut through the jargon by asking just five simple questions. I try to answer these questions as quickly as I can when skimming papers. 1) What are the inputs to the function…
Disclaimer, these are just my opinions and not necessarily those of my employer or robotics colleagues. 2021-04-23: If you liked this post, you may be interested in a more recent blog post I wrote on why I believe in end-to-end learning for robots. Hacker News Discussion Moravec's Paradox describes the observation that our AI systems can solve "adult-level cognitive" tasks like chess-playing or…
Xiaoyi Yin (尹肖贻) has kindly translated this post into Chinese ( 中文 ) Accept-or-reject decisions for the NeurIPS 2020 conference are out, with 9454 submissions and 1900 accepted papers (20% acceptance rate). Congratulations to everyone (regardless of acceptance decision) for their hard work in doing good research! It's common knowledge among machine learning (ML) researchers that acceptance…
For want of a nail the shoe was lost. For want of a shoe the horse was lost. For want of a horse the rider was lost. For want of a rider the message was lost. For want of a message the battle was lost. For want of a battle the kingdom was lost. And all for the want of a horseshoe nail. - For Want of a Nail Was the kingdom lost due to random chance? Or was it the inevitable outcome resulting from…
Xiaoyi Yin (尹肖贻) has kindly translated this post into Chinese ( 中文 ) This post was prompted by a tweet I saw from my colleague, Colin: I'm currently a researcher at Google with a "non-traditional background", where non-traditional background means "someone who doesn't have a PhD". People usually get PhDs so they can get hired for jobs that require that credential. In the case of AI/ML, this might…
A Google interview candidate recently asked me: "What are three big science questions that keep you up at night?" This was a great question because one's answer reveals so much about one's intellectual interests - here are mine: Q1: Can we imitate "thinking" from only observing behavior? Suppose you have a large fleet of autonomous vehicles with human operators driving them around diverse road…
In 1984, a panel at the AAAI conference discussed whether the field was approaching an "AI Winter" . Mitch Waldrop wrote a transcript of the discussion , and much of it reads exactly like something written 35 years into the future. Below are some quotes from the transcript that I found impressive, as they describe the feelings of many an AI researcher today and how the public views AI, despite all…
You can download a PDF (typset in LaTeX) of this blog post here . Jupyter Notebook Code on GitHub: https://github.com/ericjang/pt-jax This blog post is a tutorial on implementing path tracing, a physically-based rendering algorithm, in JAX . This code runs on the CPU, GPU, and Google Cloud TPU, and is implemented in a way that also makes it end-to-end differentiable. You can compute gradients of…
DISCLAIMER: NO INVESTMENT OR LEGAL ADVICE The Content is for informational purposes only, you should not construe any such information or other material as legal, tax, investment, financial, or other advice. Investing involves risk, please consult a financial professional before making an investment. Robinhood is a zero-commission brokerage that was founded in 2013. It has a beautiful mobile user…
JAX is a great linear algebra + automatic differentiation library for fast experimentation with and teaching machine learning. Here is a lightweight example, in just 75 lines of JAX, of how to implement Real-NVP . This post is based off of a tutorial on normalizing flows I gave at the ICML workshop on Invertible Neural Nets and Normalizing Flows . I've already written about how to implement your…
This is a tutorial on common practices in training generative models that optimize likelihood directly, such as autoregressive models and normalizing flows . Deep generative modeling is a fast-moving field, so I hope for this to be a newcomer-friendly introduction to the basic evaluation terminology used consistently across research papers, especially when it comes to modeling more complicated…
Translations: 中文 I've been at Google Brain robotics (now referred to as Robotics @ Google ) for nearly 3 years. It's helpful to reflect, from time to time, on the scientific, engineering and personal productivity takeaways gleaned from working on large research projects. Every researcher's unique experiences and experimentation can potentially become their personal competitive edge for thinking…
Snapchat's new gender-bending filter is a source of endless fun and laughs at parties. The results are very pleasing to look at. As someone who is used to working with machine learning algorithms, it's almost magical how robust this feature is. I was so duly impressed that I signed up for Snapchat and fiddled around with it this morning to try and figure out what's going on under the hood and how…
Xiaoyi Yin has graciously translated this blog post to 中文 . I often hear the remark around the proverbial AI watering hole that there are no examples of reinforcement learning (RL) deployed in commercial settings that couldn’t be replaced by simpler algorithms. This is somewhat true. If one takes RL to mean “neural networks trained with DQN / PPO / Soft-Actor Critic etc.”, then indeed, there are…
Github repo here: https://github.com/ericjang/maml-jax Adaptive behavior in humans and animals occurs at many time scales: when I use a new shower handle for the first time, it takes me a few seconds to figure out how to adjust the water temperature to my liking. Upon reading a news article, I obtain new information that I didn't have before. More difficult skills, such as mastering a musical…
Some thoughts on the interesting BagNet paper (accepted at ICLR 2019) currently being circulated around the Machine Learning Twitter Community. Disclaimer: I wasn't a reviewer of this paper for ICLR. I think it was worthy of acceptance to the conference, and hope it prompts further investigation by the research community. Please feel free to email me if you spot any mistakes / misunderstandings in…
A PDF version of this post can be found here . Chinese translation by Xiaoyi Yin Notions of uncertainty are tossed around in conversations around AI safety, risk management, portfolio optimization, scientific measurement, and insurance. Here are a few examples of colloquial use: "We want machine learning models to know what they don't know.'' "An AI responsible for diagnosing patients and…
A periodically-updated list of my favorite Deep Learning memes. Enjoy! content warning: may contain crude humor. Caption: The Gary Marcus/Yoshua Bengio debate. (Thanks Jackie Kay for sending me this)
You can find a PDF version of this blog post here . A weighted graph is a data structure consisting of some vertices and edges, and each edge has an associated cost of traversal. Let's suppose we want to compute the shortest distance from vertex $u$ to every other vertex $v$ in the graph, and we express this cost function as $\mathcal{L}_u(v)$. For example, if each edge in this graph has cost $1$,…