# Anthony W. Jung

> Machine Learning Engineer. Personal essays and research notes.

Each entry links to the original Markdown source and its canonical blog URL. Image paths starting with `/` are relative to https://anthony0727.github.io.

## Posts

- [Great Explainers](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2026-10-18-great-explainers.md): People and resources that explain things well. [Blog post](https://anthony0727.github.io/2026/10/18/great-explainers/).
- [Interesting RL papers](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2026-08-02-interesting-rl-papers.md): RL papers I find interesting. [Blog post](https://anthony0727.github.io/2026/08/01/interesting-rl-papers/).
- [Favorite research papers](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2026-04-05-favorite-papers.md): Papers of timeless value. [Blog post](https://anthony0727.github.io/2026/04/04/favorite-papers/).
- [My failure log](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-10-my-failure-log.md): It's hard to digest that I've failed. These are fragments from years past—a reflection I hope turns out to be meaningful progress. [Blog post](https://anthony0727.github.io/2025/04/10/my-failure-log/).
- [Policy gradient](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-10-policy-gradient.md): Don't ever forget about policy gradient. [Blog post](https://anthony0727.github.io/2025/04/10/policy-gradient/).
- [Function approximation and representation learning in RL](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-10-optimality-of-q.md): The discrepancy between linear RL and deep RL, and intuitive understanding of representation in RL. [Blog post](https://anthony0727.github.io/2025/04/10/optimality-of-q/).
- [From TRPO to PPO](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-13-TRPO.md): Many people use PPO without understanding it, but that's why it's such a good algorithm. [Blog post](https://anthony0727.github.io/2025/04/09/TRPO/).
- [What is the definition of existence?](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-10-what-is-existence.md): Philosophical thoughts I had when I was 13. [Blog post](https://anthony0727.github.io/2025/04/09/what-is-existence/).
- [Insights from TD (temporal difference) methods](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-10-TD.md): Sutton: “TD methods learn a guess from a guess.” [Blog post](https://anthony0727.github.io/2025/04/09/TD/).
- [State vs Observation in RL](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-09-state-vs-observation.md): I want to distinguish the words “state” and “observation”. State is what an oracle would use to fully describe the world. In a fully observed MDP, though, the agent can observe ... [Blog post](https://anthony0727.github.io/2025/04/09/state-vs-observation/).
- [Can we trust math?](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2025-04-09-can-we-trust-math.md): Thoughts I had in my early twenties. When I was just a freshman in university, I was learning (memorizing for exams actually) various axioms in mathematics, including Euclidean ... [Blog post](https://anthony0727.github.io/2025/04/09/can-we-trust-math/).
- [Overdetermined, overcomplete, and overparameterized](https://raw.githubusercontent.com/anthony0727/anthony0727.github.io/main/_posts/2024-04-10-overparameterization.md): The three terms all describe redundancy, but they compare different objects. [Blog post](https://anthony0727.github.io/2025/04/09/overparameterization/).