How to Build AI to Forecast
In this Datathon, you will play a role as a financial trader.
You will learn neural network techniques for forecasting, build your own deep learning models, and deploy your models to run live predictions of globally traded company stocks.
The top teams who generate the most profitable models will stand a chance to become AI4IMPACT FORECAST UNIVERSITY FINANCIAL FORECASTING DATATHON 2021 GLOBAL CHAMPION.
All are Welcome! No Programming Background Required to Participate.
What Will You Get?
LEARN HOW TO BUILD YOUR OWN AI FORECASTING APPS
You will learn AI concepts and apply them immediately to build actual forecasting applications. This is a great way to boost your CVs and portfolio.
FROM GLOBAL AI EXPERTS
Learning alone can be daunting. That’s why our AI Experts will directly mentor you from Day-1.
ACCESSIBLE TO NON-PROGRAMMERS
If you've never done any computer coding before, good news! We designed this bootcamp specifically for those without programming experience!
What Our Students Say
When you're getting your hands dirty, you'll learn a lot. I took an entire course about machine learning previously, but when you're actually building it out, it was a whole new learning experience for me.
Keith Lim - Singapore Management University, Quantitative Finance
Deep Learning Datathon 2020 Finalists
Your Datathon Journey
LEARN (30 August - 5 September)
Self-Paced Learning and Mentorships from AI Experts
Session 1: What is Forecasting? Key Concepts
Session 2: Risk Minimization
Session 3: The Prediction Pipeline
Session 4: Neural Networks, Multi-layer Perceptions
Session 5: Defining the Network
Session 6: Features and Neural Networks
Session 7: Training Neural Networks
Session 8: Regularization & Dropouts
DO (6 - 8 September)
Build your own financial forecasting models using real-world financial datasets. AI4IMPACT Experts provide online mentorship and masterclass.
IMPACT (9 - 11 September)
Contest (9 September)
Your models will be used to trigger virtual stock trades. The top 8 teams with the highest profits will be invited to proceed to the Finale.
Finale (11 September)
The top 8 teams will present their solutions to a panel of expert judges.
The top 3 teams will be declared as FORECAST UNIVERSITY - FINANCIAL FORECASTING 2021 CHAMPION.
Championship Certificate: for the top 3 Datathon winners
Certificate of Excellence: for the finalists who showcase great results, based on expert judges' discretions
Certificate of Completion: for all teams who beat the persistence benchmark during the contest stage
Past Students' Works
Last year, AI4IMPACT conducted a Deep Learning Datathon where 480 global participants played a role as energy traders and developed forecasting models to predict wind energy output.
Here are some of their results:
Frequently Asked Questions (FAQ)
Q: I have no knowledge of programming. Can I still participate?
Yes! The Datathon is accessible to non-programmers. No prior programming experience is needed. Learn and apply the concepts taught and work hard to get a good solution.
Q: Do I have to set up / install anything for the Datathon?
Everything will be accessible through your internet browser (Firefox or Chrome).
Q: Where is the competition held?
On the Internet :)
Q: What is the objective of the Datathon?
We want you to learn practical AI concepts and immediately apply them in building an end-to-end AI prediction application in the area of finance.
Q: Can I use external tools or frameworks during the Datathon?
No. To ensure fairness, we need to ensure the AI models submitted are their own creation and not using any ready-made models. We want all participants to experience the end-to-end process of AI development from data to deployment.
Q: How large can my team be?
You need to register in a team of 3 - 5 members. This Datathon is an intense program with lots to learn. You are strongly encouraged to sign up as a group.
Q: Can I submit multiple models for evaluation?
Each team should submit just one model for the Datathon. However, you are encouraged to tweak your model as the competition progresses.
Q: Do I need to be at a university to join?
This competition is open to anyone. Register to learn and compete.
Limited to 100 Seats
Forecasting stock prices is not a walk in the park. Although no prior programming experience is required, you need to be strongly motivated and disciplined to apply various AI techniques taught during the bootcamp.
In this bootcamp, We only accept 100 students in total, and you need to sign up in a group of 3 to 5 students. This helps us dedicate more time to personally help you build successful AI forecasts.
By pre-registering, you will receive first invitation to register on the class once the registration page is open.
Pre-Registration Deadline: 23 July 2021