Mochan Shrestha [mochan.org]
Tutorials, Education, Learning and Knowledge
RL Token: Online RL with VLAs
: An RL token is a compact, learned readout representation that compresses a …
Isaac Sim MCP Server
: Introducing isaacsim-mcp, an MCP server that lets AI coding agents (Claude, …
GAE: Generalized Advantage Estimation
: In reinforcement learning, agents learn to make better decisions by …
Weights & Biases Tutorial
: In this tutorial, we will go over how to log data to weights & biases …
MLflow Tutorial
: In this tutorial, we will go over the basics of running and logging data to …
Actor Critic Methods in Reinforcement Learning
: Actor-critic methods combine value-based learning and policy-based learning, …
Importance Sampling in Off-Policy Reinforcement Learning
: Importance sampling is a technique that solves a fundamental challenge in RL: …
Bayes Theorem Conditioned on Multiple Variables
: We want to prove Bayes theorem that is conditioned on multiple variables. We …
Reduce PDF Tool
: This is a tool to reduce the size of PDF files by removing or compressing …
Adding Perplexity.ai API support to Langchain
: We will add support for calling perplexity.ai API from LangChain. This will …
Mujoco Mocapper
: Mujoco Mocapper is a tool that takes a mujoco file and adds mocap to it. It can …
Local LLMs using Llama.Cpp and Python
: We will run local GMML models using llama.cpp and python. We will use the python …
Local LLMs using Ollama Server API with Python
: We will run local LLM models using python by calling the Ollama server API. We …
Mujoco Mocap Tutorial 2: Mujoco Mocap of the Google Robot
: We use mocap to control the Google Robot and move the end effector to a desired …
Mixed Integer Linear Programming
: Mixed integer linear programming is a set of optimization problems where some of …
Shapley Values
: Given a group of players working cooperatively, Shapley value is a measure of …
Mujoco Mocap Tutorial 1: Mocap object and keyboard movement
: In this tutorial, we will go over the mocap object in mujoco and set up a simple …
