Starting These NotesAugust 15, 2026
For a long time I told myself I’d start writing once I had something finished to write about. That bar turned out to be a moving target — so this is me starting anyway. The plan: short notes on papers I’m reading, longer derivations when a topic deserves it, and occasional blog posts. This first entry doubles as a rendering test. MathInline math works with single dollars: the TD(0) update is $V(S_t) \leftarrow V(S_t) + \alpha \left[ R_{t+1} + \gamma V(S_{t+1}) - V(S_t) \right]$. Display math uses double dollars: \[J(\theta) = \mathbb{E}_{\tau \sim \pi_\theta} \left[ \sum_{t=0}^{T} \gamma^t r(s_t, a_t) \right]\]Numbered AMS equations work too, and can be referenced: \[\begin{equation} \nabla_\theta J(\theta) = \mathbb{E}_{\pi_\theta} \left[ \nabla_\theta \log \pi_\theta(a \mid s) \, Q^{\pi_\theta}(s, a) \right] \label{eq:pg} \end{equation}\]Equation $\eqref{eq:pg}$ is the policy gradient theorem. CodeHow a new note gets addedDrop a Markdown file in |