Optimal Model Design for Reinforcement Learning

Optimal Model Design for Reinforcement Learning

This repository contains JAX code for the paper

Control-Oriented Model-Based Reinforcement Learning with Implicit Differentiation

by Evgenii Nikishin, Romina Abachi, Rishabh Agarwal, and Pierre-Luc Bacon.

Summary

Model based reinforcement learning typically trains the dynamics and reward functions by minimizing the error of predictions. The error is only a proxy to maximizing the sum of rewards, the ultimate goal of the agent, leading to the objective mismatch. We propose an end-to-end algorithm called Optimal Model Design (OMD) that optimizes the returns directly for model learning. OMD leverages the implicit function theorem to optimize the model parameters and forms the following computational graph:

Please cite our work if you find it useful in your research:

@article{nikishin2021control,
  title={Control-Oriented Model-Based Reinforcement Learning with Implicit Differentiation},
  author={Nikishin, Evgenii and Abachi, Romina and Agarwal, Rishabh and Bacon, Pierre-Luc},
  journal={arXiv preprint arXiv:2106.03273},
  year={2021}
}

Installation

We assume that you use Python 3. To install the necessary dependencies, run the following commands:

1\. virtualenv ~/env_omd
2\. source ~/env_omd/bin/activate
3\. pip install -r requirements.txt

To use JAX with GPU, follow the official instructions. To install MuJoCo, check the instructions.

Run

For historical reasons, the code is divided into 3 parts.

Tabular

All results for the tabular experiments could be reproduced by running the tabular.ipynb notebook.

To open the notebook in Google Colab, use this link.

CartPole

To train the OMD agent on CartPole, use the following commands:

cd cartpole
python train.py --agent_type omd

We also provide the implementation of the corresponding MLE and VEP baselines. To train the agents, change the --agent_type flag to mle or vep.

MuJoCo

To train the OMD agent on MuJoCo HalfCheetah-v2, use the following commands:

cd mujoco
python train.py --config.algo=omd

To train the MLE baseline, change the --config.algo flag to mle.

Acknowledgements

• Tabular experiments are based on the code from the library for fixed points in JAX
• Code for MuJoCo is based on the implementation of SAC in JAX
• Code for CartPole reuses parts of the SAC implementation in PyTorch
• For experimentation, we used a moditication of the slurm runner

Download Details:

Author: evgenii-nikishin
Download Link: Download The Source Code
Official Website: https://github.com/evgenii-nikishin/omd
License: MIT

#machine-learning #data-science

What is GEEK

Buddha Community

ML Optimization pt.1 - Gradient Descent with Python

So far in our journey through the Machine Learning universe, we covered several big topics. We investigated some regression algorithms, classification algorithms and algorithms that can be used for both types of problems (SVM**, **Decision Trees and Random Forest). Apart from that, we dipped our toes in unsupervised learning, saw how we can use this type of learning for clustering and learned about several clustering techniques.

We also talked about how to quantify machine learning model performance and how to improve it with regularization. In all these articles, we used Python for “from the scratch” implementations and libraries like TensorFlow, Pytorch and SciKit Learn. The word optimization popped out more than once in these articles, so in this and next article, we focus on optimization techniques which are an important part of the machine learning process.

In general, every machine learning algorithm is composed of three integral parts:

1. A loss function.
2. Optimization criteria based on the loss function, like a cost function.
3. Optimization technique – this process leverages training data to find a solution for optimization criteria (cost function).

As you were able to see in previous articles, some algorithms were created intuitively and didn’t have optimization criteria in mind. In fact, mathematical explanations of why and how these algorithms work were done later. Some of these algorithms are Decision Trees and kNN. Other algorithms, which were developed later had this thing in mind beforehand. SVMis one example.

During the training, we change the parameters of our machine learning model to try and minimize the loss function. However, the question of how do you change those parameters arises. Also, by how much should we change them during training and when. To answer all these questions we use optimizers. They put all different parts of the machine learning algorithm together. So far we mentioned Gradient Decent as an optimization technique, but we haven’t explored it in more detail. In this article, we focus on that and we cover the grandfather of all optimization techniques and its variation. Note that these techniques are not machine learning algorithms. They are solvers of minimization problems in which the function to minimize has a gradient in most points of its domain.

Dataset & Prerequisites

Data that we use in this article is the famous Boston Housing Dataset . This dataset is composed 14 features and contains information collected by the U.S Census Service concerning housing in the area of Boston Mass. It is a small dataset  with only 506 samples.

For the purpose of this article, make sure that you have installed the following _Python _libraries:

• **NumPy **– Follow this guide if you need help with installation.
• **SciKit Learn **– Follow this guide if you need help with installation.
• Pandas – Follow this guide if you need help with installation.

Once installed make sure that you have imported all the necessary modules that are used in this tutorial.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import SGDRegressor

Apart from that, it would be good to be at least familiar with the basics of linear algebra, calculus and probability.

Why do we use Optimizers?

Note that we also use simple Linear Regression in all examples. Due to the fact that we explore optimizationtechniques, we picked the easiest machine learning algorithm. You can see more details about Linear regression here. As a quick reminder the formula for linear regression goes like this:

where w and b are parameters of the machine learning algorithm. The entire point of the training process is to set the correct values to the w and b, so we get the desired output from the machine learning model. This means that we are trying to make the value of our error vector as small as possible, i.e. to find a global minimum of the cost function.

One way of solving this problem is to use calculus. We could compute derivatives and then use them to find places where is an extrema of the cost function. However, the cost function is not a function of one or a few variables; it is a function of all parameters of a machine learning algorithm, so these calculations will quickly grow into a monster. That is why we use these optimizers.

#ai #machine learning #python #artificaial inteligance #artificial intelligence #batch gradient descent #data science #datascience #deep learning #from scratch #gradient descent #machine learning #machine learning optimizers #ml optimization #optimizers #scikit learn #software #software craft #software craftsmanship #software development #stochastic gradient descent

Attend The Full Day Hands-On Workshop On Reinforcement Learning

The Association of Data Scientists (AdaSci), a global professional body of data science and ML practitioners, is holding a full-day workshop on building games using reinforcement learning on Saturday, February 20.

Artificial intelligence systems are outperforming humans at many tasks, starting from driving cars, recognising images and objects, generating voices to imitating art, predicting weather, playing chess etc. AlphaGo, DOTA2, StarCraft II etc are a study in reinforcement learning.

Reinforcement learning enables the agent to learn and perform a task under uncertainty in a complex environment. The machine learning paradigm is currently applied to various fields like robotics, pattern recognition, personalised medical treatment, drug discovery, speech recognition, and more.

With an increase in the exciting applications of reinforcement learning across the industries, the demand for RL experts has soared. Taking the cue, the Association of Data Scientists, in collaboration with Analytics India Magazine, is bringing an extensive workshop on reinforcement learning aimed at developers and machine learning practitioners.

#ai workshops #deep reinforcement learning workshop #future of deep reinforcement learning #reinforcement learning #workshop on a saturday #workshop on deep reinforcement learning

Penalizing the Discount Factor in Reinforcement Learning

This post deals with the key parameter I found as a high influence: the discount factor. It discusses the time-based penalization to achieve better performances, where discount factor is modified accordingly.

I assume that if you land on this post, you are already familiar with the RL terminology. If it is not the case, then I highly recommend these blogs which provide a great background, before you continue: Intro1 and Intro2.

What is the role of the discount factor in RL?

The discount factor, **𝛾, **is a real value ∈ [0, 1], cares for the rewards agent achieved in the past, present, and future. In different words, it relates the rewards to the time domain. Let’s explore the two following cases:

1. If **𝛾 **= 0, the agent cares for his first reward only.
2. If **𝛾 **= 1, the agent cares for all future rewards.

Generally, the designer should predefine the discount factor for the scenario episode. This might raise many stability problems and can be ended without achieving the desired goal. However, by exploring some parameters many problems can be solved with converged solutions. For further reading on the discount factor and the rule of thumb for selecting it for robotics applications, I recommend reading: resource3.

#artificial-intelligence #optimization #deep-learning #machine-learning #reinforcement-learning #deep learning

Workshop Alert! Deep Learning Model Deployment & Management

The Association of Data Scientists (AdaSci), the premier global professional body of data science and ML practitioners, has announced a hands-on workshop on deep learning model deployment on February 6, Saturday.

Over the last few years, the applications of deep learning models have increased exponentially, with use cases ranging from automated driving, fraud detection, healthcare, voice assistants, machine translation and text generation.

Typically, when data scientists start machine learning model development, they mostly focus on the algorithms to use, feature engineering process, and hyperparameters to make the model more accurate. However, model deployment is the most critical step in the machine learning pipeline. As a matter of fact, models can only be beneficial to a business if deployed and managed correctly. Model deployment or management is probably the most under discussed topic.

In this workshop, the attendees get to learn about ML lifecycle, from gathering data to the deployment of models. Researchers and data scientists can build a pipeline to log and deploy machine learning models. Alongside, they will be able to learn about the challenges associated with machine learning models in production and handling different toolkits to track and monitor these models once deployed.

#hands on deep learning #machine learning model deployment #machine learning models #model deployment #model deployment workshop

Intro to Reinforcement Learning: Temporal Difference Learning, SARSA Vs. Q-learning

Reinforcement learning (RL) is surely a rising field, with the huge influence from the performance of AlphaZero (the best chess engine as of now). RL is a subfield of machine learning that teaches agents to perform in an environment to maximize rewards overtime.

Among RL’s model-free methods is temporal difference (TD) learning, with SARSA and Q-learning (QL) being two of the most used algorithms. I chose to explore SARSA and QL to highlight a subtle difference between on-policy learning and off-learning, which we will discuss later in the post.

This post assumes you have basic knowledge of the agent, environment, action, and rewards within RL’s scope. A brief introduction can be found here.

The outline of this post include:

• Temporal difference learning (TD learning)
• Parameters
• QL & SARSA
• Comparison
• Implementation
• Conclusion

We will compare these two algorithms via the CartPole game implementation. This post’s code can be found here :QL code ,SARSA code , and the fully functioning code . (the fully-functioning code has both algorithms implemented and trained on cart pole game)

The TD learning will be a bit mathematical, but feel free to skim through and jump directly to QL and SARSA.

#reinforcement-learning #artificial-intelligence #machine-learning #deep-learning #learning