Almost Year End
I’ve noticed, especially during my master’s courses, that no matter how hard something seems at the beginning, it feels different once I start making progress.
This is my first try at blogging. Rather than telling a clear story about one topic, I’m going to take stock of where things are.
Work - OCBC AI Lab
Last year, when I was attending Nuno’s class on negotiation, one of the projects was “10 No’s.” You had to get ten rejections for the same task. Nuno said we would soon find out how hard it was to get repeatedly rejected. That project eventually landed me in OCBC AI Lab.
I didn’t take a break: I left Wego on 31 July and joined OCBC three days later. There are pros and cons to both places, and you can never know them all beforehand. These days I can focus on machine learning projects without spending time on BI work like dashboards, but I no longer have the freedom to do everything in the cloud.
Progress in the AI Lab isn’t linear. It’s more like trying until you find all the dead ends, then starting again. It’s quite an experience. I guess that’s why they hire PhDs; they seem more used to this back and forth. The key is to keep trying.
These are the projects I’m currently working on:
| Project | Difficulty | Start Date | Progress | Note |
|---|---|---|---|---|
| Enterprise Chatbot | Hard | Aug’18 | Deployment | Luis.ai & DialogFlow |
| Card Fraud | Hard | Jul’19 | Phase 2 | Anomaly Detection |
| Cheque OCR | Hard | Sep’19 | POC | CNN + RNN |
Study - Georgia Tech - M.S. CS
During my last days at Wego, apart from Nuno’s class, I also got my GMAT score: 700. It wasn’t that high, but it was enough to start applying to schools. In the end, I applied to several part-time master’s programs rather than an MBA. I chose the Master of Science in Computer Science at Georgia Tech because the other programs seemed more interested in charging high fees. OMSCS looked like the most affordable choice, and the reviews of its quality were good.
I decided on the Machine Learning specialization, a natural choice since I’ve been doing it for years. I’ll write about individual courses in future posts. Here is the general plan:
| Course | Difficulty | Semester |
|---|---|---|
| CS7638 Artificial Intelligence for Robotics | Hard | 19’Spring |
| CS7646 Machine Learning for Trading | Medium | 19’Summer |
| CS6476 Computer Vision | Hard | 19’Fall |
| CS6300 Software Development Process | Easy | 19’Fall |
Side tracks
I finished most of the Coursera specializations that interested me while I was at Wego. Looking back, it was a bit too much. This year I stopped taking Coursera courses because I’d already started my master’s. I did, however, go back to school for CS6101 for two consecutive semesters. Both were really difficult.
- First semester (18’Fall): NLP following Stanford’s CS224N.
- Second semester (19’Spring): Deep Reinforcement Learning following UC Berkeley’s CS285.
- Subsequently (19’Summer), I was granted CES to take Udacity’s Deep Learning Nanodegree.
Others
I’m still making time for the gym and exercise. Since injuring my ankle, though, I’ve skipped most of my regular soccer and basketball games.
This year I finally decided to resume piano lessons. Conveniently, there’s a piano school 50 metres from my place. The teacher suggested ABRSM grade 7, but I felt too tired to make another commitment. Nothing can be enjoyable anymore if you put an exam on it.
These are the two pieces I would like to master next year:
- Piano Sonata No. 8 in C minor, Op. 13 III. Rondo: Allegro
- Prelude and Fugue in C minor, BWV 847
By the end of next year, hopefully I can start these two:
- Chopin: Fantaisie-impromptu in C-Sharp Minor, Op. 66
- Beethoven: The Piano Sonata No. 14 in C♯ minor “Quasi una fantasia”, Op. 27, No. 2
I believe you should stop now and then to think about long-term goals. Next year’s priorities will still be work and study. Beyond that, instead of taking more side courses, I’d like to get some Kaggle exposure and do some tech sharing. Replicating state-of-the-art results and writing about them on Medium would be a good start. Eventually, I’d also like to try a publication. I may need some help from a professor with that.
Data science can be an exhausting journey. I believe the best way through is to keep it fun.
Satisfaction = Perception - Expectation
Travel
I need a separate section for travel since I’ve been spending too much on it.
Travel has always played a big part in my life. It’s a beautiful world, and I want to see it with my own eyes.
| Time | Places | Sub-places | Score |
|---|---|---|---|
| Dec’18 | China | Beijing, Pingyao, Xi’An | ⭐⭐ |
| Jan’19 | China | Dalian | ⭐⭐⭐⭐⭐ |
| May’19 | Japan | Kobe, Himeji, Kyoto, Osaka | ⭐⭐⭐⭐ |
| Aug’19 | Central Europe | Germany, Czech, Slovenia, Croatia, Montenegro, Hungary, Slovakia | ⭐⭐⭐⭐⭐ |
| Oct’19 | Myanmar | Yangon, Bagan | ⭐⭐⭐⭐ |
| Dec’19 | China, Korea | Shenzhen, Chengdu, Chongqing, Ningbo, Seoul, Dalian | ⭐⭐⭐ |