#002 | My BEST Morning Routine

Hi, I’m Ryan! You are receiving this because you have signed up to my weekly newsletter for Natural Language Processing (#NLP365), Entrepreneurship, and Life Design content!

Jan 17, 2021

Hey friends,

This is the 2nd week of 2021. Since last Monday, I have kickstarted a new morning routine (due to covid lockdown) that has worked out surprisingly well! It goes as follows:

5am : Wake up
5am - 5:30am : Mentally wake up + get myself ready to workout + weigh myself
5:30am - 6am : Workout (200 reps of a main exercise + rear delt lateral raise)
6am - 7am : “Read” (Audible)
7am - 7:30am : Make notes on what I read

…and then my days start, either meetings at 7:30am or shower + breakfast to start work at 9am. This routine has allowed me to read 5 books in the first two weeks of 2021. To put into perspective, I read a total of 4 books last year 🤣

For years I have tried to read in the morning but it just ends up making me feel very sleepy… With this morning routine, not only did I work on both my body and my mind but more importantly, I feel SOOO GOOOD afterwards, having accomplished so much and it’s usually still dark outside! This morning routine has successfully given me a strong momentum boost everyday since I started it and I am not sure if I can go back to my old routine of gym from 6am - 7:15am! 😬

This week I finished reading:

The Lean Startup (8th Jan - 12th Jan 2021)
Make Time (12th Jan - 14th Jan 2021)
Hooked (15th Jan - 16th Jan 2021)

Total: 5 / 26 books | 0 / 26 level 4 notes | 0 / 12 actions

❓Question of the Week

What’s your morning routine?

Morning routine is so important as it sets you up for the day. Have a bad morning and your whole day sucks. Have a good morning and you thrive throughout the day.

The good thing about morning routine is that it doesn’t have to be complicated. It just has to work for you. A good morning routine is one that really gives you lots of positive energy and momentum for you to conquer the day and take on whatever comes your way throughout the day.

The question now is how do you find YOUR ideal morning routine?

Hypothesis / Plan —> Execute —> Reflect —> Iterate.

In fact, you can use this 4 stage workflow to find what suits you best in almost anything in life.

🐦 Tweet of the Week

Content Philosopher ☕ @content_wisdom

"The tragedy is that much of what you think is random is in your control and, what’s worse, the opposite." @nntaleb

💡 Quote of the Week

Triggers come in two types — external and internal. External triggers tell the user what to do next by placing information within the user’s environment. Internal triggers tell the user what to do next through associations stored in the user’s memory — Hooked

🔥 Recommendation(s) of the Week

Sleep Cycle — I have been using this app as my alarm for almost 2 years now. The beauty of this app is that it tracks my snoring… wait what??? Nah just kidding 🙄

The beauty of this app is that a) it aims to wake you up when you are not in the deep sleep stage and b) the alarm goes from low volume to high volume gradually. Both of these things is to prevent you from waking up feeling groggy. It has worked out pretty well for me over the last 2 years!

🔦 AI Research Papers Spotlight of the Week

Rethinking Generalization of Neural Models: A Named Entity Recognition Case Study

In this paper, they aimed to diagnose and characterise generalisation in the context of Named Entity Recognition. They introduced 4 measures and ran 6 experiments in order to answer the following three research questions:

Does our model really have generalisation ability?
What factor of a dataset can distinguish neural networks that generalise well from those that don’t?
How does the relationship between entity categories influence the difficulty of model learning?

Improving Multimodal Named Entity Recognition via Entity Span Detection with Unified Multimodal Transformer

The paper proposed Multimodal Transformer for MNER, which combine Transformer with a multimodal interaction module to capture the inter-modality dynamics between words and images. There are three problems with existing work in MNER:

Non-contextualised word representations
Word representations in the final hidden layer are still based on textual context which are insensitive to the visual context
Largely ignore the bias of visual information. Most visual only highlights one or two entities in the sentence

🎥 This Week on YouTube

This week’s video is Episode #004 of Research Papers Summary, covering Temporally-Informed Analysis of NER 🎮

That’s it for this week! I hope you find something useful from this newsletter. More to come next Sunday! Have a good week ahead! 🎮

More of me on YouTube, Twitter, LinkedIn, and Instagram.

Actionable Knowledge by Ryan Ong