Restrict function search space by assumptions

Gaussian process

In the previous post, I discussed the function fitting view of supervised learning. It is theoretically impossible to find the best fitting function from an infinite search space. In this post, I will discuss how we can restrict the search space in function fitting with assumptions. 

Bento for your mind while commuting on the grind

My commute to work is a 3-hour journey every day, with 1 hour on the train and half an hour on foot each way. Friends who visit me from the city always ask, “Do you commute like this every day? Aren’t you exhausted?” I thought this way too before moving out from the city, where I walked for only 10 minutes between home and school. But now, I am glad that I have 3 full hours of uninterrupted solo time every day, and the switch of environment between the fast-paced urban and the serene suburban life helps refresh my brain and reset my mind.

I prepare small “learning bento” before going to work, and spend the time on the train either watching MOOC (such as Coursera) and YouTube videos, or reading books and articles, and spend the time on foot listening to podcast or Audible. 3 hours a day adds up to 15 hours a week and 60 hours a month. And over the past few months, I have discovered a lot of great learning resources and would like to share with those who also like to have some “learning bento” for the mind while commuting on the grind.

How NOT to get a job offer as new grads

Disclaimer: This post represents my personal experience only, and has nothing to do with any organization or group. 

When I was a student, I read tons of articles on how to prepare for job hunting and how to get a job offer. Now, I am sitting on the other side of the table and start to participate in the talent acquisition process myself. This gives me a different perspective on the recruiting process. In particular, how I decide NOT to move forward with a candidate.  Here I am giving a few examples for new grads: how NOT to get a job offer. Follow this list, and you will likely NOT get an offer. So, try not to be a follower.

Non-linearity and local optimum mindset

Time flies. It’s been 7 months since I started to work as a full-time Data Scientist. It sometimes feels much shorter than 7 months: imaging neurons in the laboratory as a graduate student and walking on the stairs in front of Alma mater on campus was just like yesterday. It sometimes feels much longer than that: working is so vastly different from academic study, and I’ve learned so much on so many aspects of data science within this short-long period of time, that my mind and understanding of the industry and the world hardly resembles the graduate student me.

Here I am summarizing a few lessons I learned from work.

Not ready yet…Go!

I officially graduated from Columbia last Tuesday after my PhD thesis defense, and started to work full time immediately in the following week. This is the first week of my transition from academia to industry, and I already experience some culture shock in time management and prioritizing. Learning on the job is not only about the technical details, but also about the pattern of thinking.