Sitemap

A list of all the posts and pages found on the site. For you robots out there, there is an XML version available for digesting as well.

Pages

Posts

portfolio

Flappy-Bird-AI

A PPO agent that learns to play Flappy Bird; a small sandbox for reward shaping and observation design.

Gradescope MCP

An MCP server with 74 tools that lets AI assistants work through real Gradescope grading: assignments, rubrics, submissions, extensions and regrade requests. Writes to Gradescope return a preview first and run only on an explicit confirmation, which the agent is told to get from a person.

Run-Skills

An agent skill, with a small harness, that lets Claude Code, Codex and pi run experiments on Slurm clusters and GPU workstations. The agent syncs the project, builds its environment there, checks each launch against what you have already allowed, and watches the run through small, bounded reads.

RunFleet

A terminal dashboard for Slurm jobs and GPUs across every cluster and workstation you can ssh into, with nothing installed on the remote side. Jobs that ended get a likely cause, linked to the log line.

StudyGround

A Claude Code plugin for self-study: lessons in a local web reader, Python that runs in the browser through Pyodide, a course-aware tutor, and exercises that open in VS Code.

publications

TOPPO: Rethinking PPO for Multi-Task Reinforcement Learning with Critic Balancing

NeurIPS 2026, poster

In multi-task PPO, the critic’s per-task gradients differ in scale far more than the actor’s, and the hardest tasks stall. Balancing the critic gives a 717K-parameter shared actor-critic the strongest tail on Meta-World+ MT50: 56.5% success on each seed’s ten worst tasks, against 43.3% for ARS-LN-1024, a baseline with 22.7× the parameters, rerun under the same protocol.

Yuanpeng Li, Gefei Lin, Annie Qu, Rui Miao. In Advances in Neural Information Processing Systems (NeurIPS), 2026. arXiv:2605.11473

talks

teaching

STATS 205P — Bayesian Data Analysis

Graduate course, UC Irvine, Master of Data Science Program, 2024

Graduate teaching assistant. Prior specification, Bayesian linear and generalized linear regression for the Master of Data Science program.

STATS 7 — Basic Statistics

Undergraduate course, UC Irvine, Department of Statistics, 2024

Graduate teaching assistant. Introductory inferential statistics: confidence intervals, hypothesis testing, and regression.