The CCF EconCS annual gathering is taking place at Nanjing University this year. The Associate Head of Jing’an District delivered an opening talk on how big data and related technologies are driving economic growth. About Jing’an District 静安 JingAn district is in central Shanghai. It is the most fancy commercial area — aka shopping heaven. Like, it is the place I take my friends to if they’re visiting Shanghai everytime (like I’ve been there too many times during my undergrad)....

Tchaikovsky’s Sleeping Beauty is one of the few works he was genuinely proud of — which is remarkable, considering his famously self-critic personality. The score is so stunningly good that, at times, the ballet choreography seems boring compared to the music. Tchaikovsky was notorious for writing ballet music so difficult that dancers struggled to find tempo and balance on stage. (Of course, critics in 1890 hadn’t yet met Stravinsky’s Rite of Spring — that riot was still 40 years away....

Bizet’s Carmen has always held a special place in my heart. As a young flutist in the orchestra, I was enchanted by its endless flute solos — each line carrying those unforgettable melodies. Among them, the “Intermezzo II” from the suite remains one of the most charming pieces I’ve ever played: And when our orchestra performed Fantaisie Brillante sur Carmen, I watched from the second flute chair as the principal soared through the piece....

A Bilibili influencer recently posted a scathing critique of a major Chinese streaming app’s advertising practices — calling out the flood of ads, shady tactics, and outright ad fraud. Takeaways Ads on logos — yes, really Some apps even slap ads directly on their logos just to squeeze out more exposure. Ewww. These logo-banner ads even show up in the App Store. Imagine Apple spending decades perfecting its ecosystem, only to have Chinese apps free-riding on their storefront real estate....

There’s a lovely scene at the end of Étoile (2025) (Amazon’s TV about ballet), when the boy went to watched the first ballet show and then waited backstage for the ballerina for a date: ‘There were a lot of feathers up there… And I (almost) didn’t fall asleep! The boy said after his first ballet, probably swan lake “There are a lot of feathers. I fall asleep once — but not when you’re on stage… And why a swan?...

Visualization of the Mitosis algorithm. Here’s a poster for the (preliminary version of) the paper Contextual Budget Bandit for Food Rescue. Still working on some last 20% works that takes 80% of the time before pushing it to ArXiv.

I remember once going on a date when my friend casually said, “I love opera. Someday, I want to go to London and watch The Phantom of the Opera live.” At that moment, I was honestly terrified — what an ignorant, pretentious idiot?! But then… I stumbled across a live concert by The 3 Tenors — Plácido Domingo, José Carreras, and Luciano Pavarotti — accompanied by the Los Angeles Philharmonic Orchestra, conducted by the legendary Zubin Mehta....

A lighthearted vignette.

As I’m practicing this piece, here’s a bit more about it: “In my concertos I have allotted the instruments the types of themes which correspond to their particular tone qualities and respect their expressive possibilities.” This statement of Jacques Ibert’s certainly applies to his Flute Concerto, written over the years 1932-1933. The work was dedicated to Marcel Moyse (who’s quite the central spotlight of Paris-school flute, read more about him here), who was the featured soloist in its premiere performance, under Philippe Gaubert’s direction, in Paris on February 25, 1934....

Here’s one more interesting matrix completion paper: Matrix Completion from Non-Uniformly Sampled Entries Yuanyu Wan, Jinfeng Yi, Lijun Zhang https://doi.org/10.48550/arXiv.1806.10308 we assume that a small number of columns are randomly selected and fully observed, and each remaining column is partially observed with uniform sampling. To recover the unknown matrix, we first recover its column space from the fully observed columns. Then, for each partially observed column, we recover it by finding a vector which lies in the recovered column space and consists of the observed entries....