From a553a68dbe85df21f3d5749cbc309f4c5c26d0a8 Mon Sep 17 00:00:00 2001
From: Harshit Sikchi <hari-sikchi@users.noreply.github.com>
Date: Fri, 29 Mar 2019 15:57:22 +0530
Subject: [PATCH 1/3] Update README.md

---
 README.md | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/README.md b/README.md
index 1957316..6478383 100644
--- a/README.md
+++ b/README.md
@@ -5,6 +5,7 @@
    * [HCOPE](#hcope)
    * [Safe Exploration](#safe_exploration)
    * [Off Policy Evaluation](#importance_sampling)
+   * [Solving side effects](#side_effects)
 <!--te-->
 
 
@@ -117,3 +118,11 @@ Comparision of different importance sampling estimators:
  Image is taken from phD thesis of P.Thomas:    
  Links: https://people.cs.umass.edu/~pthomas/papers/Thomas2015c.pdf   
 
+
+   
+## Penalizing side effects using relative reachability
+
+* Added a simple example for calculating side effects as given towards the end of paper
+
+    
+ Paper: Penalizing side effects using stepwise relative reachability - Krakovna et al.

From 90347be808124147c8191522292e65fdadd86adb Mon Sep 17 00:00:00 2001
From: Harshit Sikchi <hari-sikchi@users.noreply.github.com>
Date: Sat, 30 Mar 2019 13:09:27 +0530
Subject: [PATCH 2/3] Update README.md

---
 README.md | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index 6478383..75aed1d 100644
--- a/README.md
+++ b/README.md
@@ -9,7 +9,7 @@
 <!--te-->
 
 
-
+***
 
 <a name="hcope"></a>
 ## HCOPE (High-Confidence Off-Policy Evaluation.)
@@ -55,6 +55,8 @@ method. Also, a graph of distribution of Importance sampling ratio is created wh
 Output format: 
 ![Output](https://github.com/hari-sikchi/safeRL/blob/master/results/Result.png)   
 
+***
+
 <a name="safe_exploration"></a>
 ## Safe exploration in continuous action spaces.
 
@@ -99,7 +101,7 @@ This enables agent to learn while following the safety constraints.
 ![Action Correction](https://github.com/hari-sikchi/safeRL/blob/master/results/safety_optimization.png)   
 
 
-
+***
 
 <a name="importance_sampling"></a>
 ## Importance Sampling
@@ -119,10 +121,24 @@ Comparision of different importance sampling estimators:
  Links: https://people.cs.umass.edu/~pthomas/papers/Thomas2015c.pdf   
 
 
-   
-## Penalizing side effects using relative reachability
+<a name="side_effects"></a>
+
+
+*** 
+ 
+## Side Effects
+### Penalizing side effects using relative reachability   
+
+Code - https://github.com/hari-sikchi/safeRL/tree/safe_recovery/side_effects    
+
 
 * Added a simple example for calculating side effects as given towards the end of paper
+![Environment](https://github.com/hari-sikchi/safeRL/blob/safe_recovery/side_effects/env.png)   
+
+The relative reachability measure    
+![Equation relative reachability](https://github.com/hari-sikchi/safeRL/blob/safe_recovery/side_effects/rr.png)   
+
+
 
     
  Paper: Penalizing side effects using stepwise relative reachability - Krakovna et al.

From 35388b86b4a645ec7870421404d114d70821a333 Mon Sep 17 00:00:00 2001
From: Harshit Sikchi <hari-sikchi@users.noreply.github.com>
Date: Sat, 30 Mar 2019 13:10:49 +0530
Subject: [PATCH 3/3] Update README.md

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

diff --git a/README.md b/README.md
index 75aed1d..03598c3 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,6 @@
 # Safe Reinforcement Learning Algorithms   
-=================
+
+***
 
 <!--ts-->
    * [HCOPE](#hcope)