You Can’t Miss these 4 Powerful Reinforcement Learning Sessions at DataHack Summit 2019
“If intelligence was a cake, unsupervised learning would be the cake, supervised learning would be the icing on the cake, and reinforcement learning would be the cherry on the cake.” – Yann LeCun
I love that analogy! Reinforcement learning is an intriguing and complex field. We at Analytics Vidhya are strongly behind the incredible potential of this domain and the breakthroughs and research by behemoths like DeepMind support our thought process.
However, there is a general perception in the community that RL’s use cases are limited to computer simulations. This is a bit of a myth. Reinforcement learning has a number of applications in the industry! Check out this quite from Professor Balaraman Ravindran, an RL expert, and a speaker at DataHack Summit 2018:
In this article, I am thrilled to present four amazing sessions at DataHack Summit 2019, India’s largest applied Artificial Intelligence and Machine Learning conference! You get to hear from and interact with eminent reinforcement learning experts and practitioners who are currently working on real-world RL use cases.
This is an opportunity you really don’t want to miss! There are very few seats left so I recommend reserving your spot today.
I also encourage you to take some time and listen to these two insightful DataHack Radio podcast episodes on reinforcement learning:
- Reinforcement Learning with Professor Balaraman Ravindran
Here are the Powerful Reinforcement Learning Sessions at DataHack Summit 2019
- Power Talk: Finding the Shortest Route for a Cab Ride using Reinforcement Learning by Sayan Ranu
- Power Talk: Reinforcement Learning in the Real World: An Industrial Applications Perspective by Dr. Harshad Khadilkar
- Hack Session: Reinforcement Learning in Real Life – Implementation Guide by Hardik Meisheri and Richa Verma
- Hack Session: Automated Portfolio Management using Reinforcement Learning by Sonam Srivastava
If you’re new to the world of Reinforcement Learning or are looking to brush up your skills ahead of DataHack Summit 2019, I highly recommend checking out our collection of articles here and an in-depth beginner guide here.
I’m sure you’ve all used the Uber or Ola (or similar service) platform to hail a ride. We’ve all become quite reliant on these cab aggregators to get us from one place to another in double-quick time.
One of the most underappreciated aspects, from a customer perspective, is the short waiting time between booking a cab and the ride arriving at your doorstep. Think about it – would you wait half an hour for a cab? Unlikely, right?
In today’s highly competitive taxi service industry, anticipating the location of future customer requests and selecting routes accordingly is critical toward gaining a competitive advantage.
Such strategically selected routes lead to shorter wait times for customers and reduced fuel costs for taxi drivers. A win-win for everyone!
In this Power Talk, Dr. Sayan Ranu, Assistant Professor at IIT Delhi, will discuss the algorithms to achieve this end goal. Through extensive empirical evaluation on real datasets, he will present evidence that the proposed strategies lead to up to 70% shorter waiting times for customers, 40% more customers, and a 20% lower rejection rate.
Key Takeaways from this Power Talk:
- An overview of route recommendation and how it works
- Understanding the impact of reinforcement learning on the cab industry
This is a talk you don’t want to miss!
This is one of the most highly anticipated Power Talks at DataHack Summit 2019. How do you bring research developments in reinforcement learning into the real world? Where and how does reinforcement learning (RL) work in an industry setting?
RL has held a lot of promise for a number of years now (it’s not a coincidence that DeepMind has gone all in on RL). But progress has been slow in terms of getting it into the industry.
Here’s the good news – there are experts who have been able to bridge this gap successfully. And I’m thrilled to announce one such expert, Dr. Harshad Khadilkar, who’ll be talking about the real-world applications of Reinforcement Learning.
He will give a quick overview of reinforcement learning and its internal working, sticking to mathematical intuition rather than rigorous equations and derivations.
Following this, he will highlight key challenges folks face when moving reinforcement learning from computer simulations and games into the real world.
The talk will focus on potential solutions to these problems, with practical examples from the fields of transportation, logistics, and supply chain operations.
Here’s a quick video for you to discover the key takeaways from Dr. Harshad himself:
Another practical reason why you just cannot miss being at DHS 2019! So reserve your seat and be there!
We’ve heard about how to bridge the gap between using reinforcement learning in a simulation setting and getting into an industrial application. It’s time to see how the code for that works!
I’m sure this will catch the fancy of all data scientists and machine learning practitioners. This is a rare opportunity to see a hands-on coding session on an industry-level reinforcement learning topic.
Here are the key topics Richa and Hardik will be presenting in this hack session:
- Brief introduction of the application of RL on real-life problems and not gaming environments
- Walkthrough of the code of applying RL over simple use cases, such as solving the online 3D bin packing problem
- The major focus would be on designing architecture, training paradigm and structuring experiments
And here’s a short video by Hardik Meisheri on what you can expect from this hack session:
Hack sessions are one-hour hands-on sessions on trending case studies & applications in machine learning, deep learning, reinforcement learning, NLP, and more.
Finance just seems like the perfect domain for machine learning, right? It’s all about numbers, finding patterns, analyzing trends, and so on. But where does reinforcement learning come into the picture?
Well, we’re about to find out in this fascinating hack session by Sonam Srivastava! Sonam will be taking a hands-on coding session to showcase how we can use reinforcement learning to automate portfolio management.
The applications of reinforcement learning in finance are still nascent but the potential is undoubtedly unparalleled. Here is the structure of Sonam’s hack session:
- Introduction to deep reinforcement learning and how to define an RL problem?
- Introduction to the problem statement and definition of the network architecture
- Types of networks used
- Scoring functions
- How to optimize for costs and other nuances
- The demo of the notebook:
- Exploratory data analysis
- Demonstration of the RL framework + other comparative frameworks
- Demonstration of results in comparison with other comparative frameworks
- Resources/Papers to find more about deep RL for portfolio optimization
Here are more insights on the key takeaways from Sonam’s hack session:
So, what are you waiting for?
Reinforcement learning is shaping up to be a massive industry in the near future (if it isn’t already!). You will not get many chances to attend live talks and watch industry-level code from RL experts. This is a wonderful opportunity to upskill your existing portfolio.
So happy reinforcement learning! And if you have any questions on RL or any of the above talks, let’s discuss that in the comments section below.