From 866bd2b5b0c18ce1ae2a91e30ae8de6844cb730b Mon Sep 17 00:00:00 2001
From: fczyzyk <112985133+fczyzyk@users.noreply.github.com>
Date: Tue, 15 Oct 2024 11:09:28 -0400
Subject: [PATCH] Update dp.md

---
 dp.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/dp.md b/dp.md
index 72bd1cce..d8e395dc 100644
--- a/dp.md
+++ b/dp.md
@@ -811,8 +811,9 @@ $$
 More often, we refer to $v_\gamma$ by simply calling it the optimal value function. 
 
 As for the finite-horizon setting, the infinite horizon discounted model does not require history-dependent policies, since for any $\pi \in \Pi^{HR}$ there exists a $\pi^{\prime} \in \Pi^{MR}$ with identical total discounted reward:
+
 $$
-v_\gamma^*(s) \equiv \max_{\pi \in \Pi^{HR}} v_\gamma^\pi(s)=\max_{\pi \in \Pi^{MR}} v_\gamma^\pi(s) .
+v_\gamma^*(s) \equiv \max_{\pi \in \Pi^{HR}} v_\gamma^\pi(s)=\max_{\pi \in \Pi^{MR}} v_\gamma^\pi(s)
 $$
 
 ## Random Horizon Interpretation of Discounting