View Entry

Reinforcement Learning for Financial Sector- DRAFT 

28 days ago

Business Value and Use Case

The Financial industry has been exploring the use of AI and machine learning in their use-cases, but given the monetary risk associated with the transactions and implementation, they post resistance. Traditional algorithmic trading has evolved in recent years and now high-computational systems automates the tasks, but traders still build the policies that govern choices to buy and sell. An algorithmic model for buying stocks based on a list of valuation and growth metric conditions might define a “buy” or “sell” signal that would in turn be triggered by some specific rules that the trader has defined. Compared to traditional machine learning algorithms, using reinforcement learning makes the entire machine trading a fully automated method, as the policy building is now done by the model and not the traders. Using RL, the reward is hypothesized to be better than the policies manually built by traders, as hundreds of signals can be much efficiently analyzed by algorithms.

Assets Included

Prerequisites of Cloud Pak for Data Services 

- NA

Enablement Material


Import the Accelerator

- Go to scripts/Final_Agent to see the RL model.
- Go to scripts/Final_Functions to see all the utility function for the accelerator.
- Go to scripts/Final_Train to train the RL model.
- Go to scripts/Final_Evaluate to see the code used to evaluate the RL model on the new stock price data

Next Steps

Aishwarya Srinivasan (

Release Notes

- information of version the accelerator has been tested with.

About the Developer


Importing the accelerator


0 Favorited
0 Files