angela zhou
@angelamczhou
assistant prof at USC Data Sciences and Operations and Computer Science; phd Cornell ORIE. data-driven decision-making, operations research/management, causal inference, algorithmic fairness/equity bureaucratic justice warrior angelamzhou.github.io
Man I got so much stuff to do I dont have the time to write a "Eugenics is bad. Factor analysis is useful sometimes. We shouldn't write papers that celebrate eugenics in the second paragraph, even if the rest of the paper is about something else"
This is a bit of a "hard launch" for our series of workshops and whitepapers with a multidisciplinary crew of amazing folks! Longer paper: arxiv.org/abs/2507.05216 Shorter perspective: arxiv.org/abs/2606.25668 Your predictive ADS is actually an organizational process, decision, and change problem
Check out our #FAccT2026 tutorial tomorrow on Bridging Predictions and Interventions in Social Systems! We're going to be building a predictions - interventions index to track ADS systems, evaluations, and gaps therein. Bring a device; I'll try to bring worksheets too :)
Agencies already know they need more staff, but we show how our richer model enables *bundled interventions* that combine extra staff with service design changes. For example, DSS estimated it needed 150 staff to get waits < 2 min: if they could reduce handling time by 50%, they only need 50 more.
We developed a queueing model to support performance evaluation and improvement snap-call-center-access.org Current staffing guidance is overly optimistic because when callers can't get through to complete an interview in 30 days, they'll call back - this *endogenous congestion* worsens backlogs
im just a rat that types and writes papers (cough revises) at verve coffee
I just submitted a paper that it took me three years to expand from the original conf submission and it feels like I wrenched another being out of my body and on the way out it turned myself inside out
vibe coded a prototype here github.com/angelamzhou/... goals: semantic edits to matplotlib, edit latex w/ preview, surface edits in code. Still barebones and mostly untested. but nice to be able to spin up bespoke solutions for oneself (even if they're janky shims for currently flawed processes)
Its nice to make stuff Made a lil parallax explainer w claude from an existing tikz and scribbles
Claude itself works well - but all setups suffer from limited visibility through the messaging interface (eg if an auth window opens onscreen)
i splurged on an eink monitor to crank through my backlog of projects to ship out the door. it's great. every day just trying to get closer to a liberation from screens
everything is bad, but my maesil chong -> moonshine maesil soju is good www.maangchi.com/recipe/maesi... it's especially useful as a marker of time, the maesil went into sugar around last may it's strong, but if i put the maesil cheong syrup back into the soju and dilute it tastes like choya
Staging down the street from me in between the boba and noodle shop ...... we are so far down the bad timeline
Started my maesil cheong after some may deadlines www.maangchi.com/recipe/maesi... the syrup will be ready for next semester And the plum wine will be ready for next year's deadlines 🫠
from the nyt article on project esther - why building solidarity is so important right now