Make the Earth a better place with machine learning at re:Invent 2021
Learn how to fight climate change with machine learning skills at one of the biggest tech conferences
From time to time, I contemplate how lucky I am to do a work I really enjoy. I suppose most machine learning practitioners agree with me: we love doing our job because it is fun, interesting, challenging, and ever-changing. Still, sometimes I stop and think about: besides the intellectual satisfaction, can I be proud of what am I doing also in the moral sense? Do I make the world better with my work?
Without a doubt, one of our and the coming generation’s biggest issues is climate change. We write notices in public toilets saying, “Please leave this place at least as tidy as you’ve found it!” but we can’t keep up with this simple message regarding our planet. So, what if you could utilize your machine learning skills to make the Earth a better place? The good news is that if you want to learn about how to do it, you should just tune in to the following sessions on AWS re:Invent.
By the way, what is AWS re:Invent? It is the one of the most influential annual learning conferences in the cloud space. Organized by Amazon Web Services in 2021 it will happen between November 29 and December 3 in Las Vegas. If you can’t participate in person, most of the learning materials will be available online for free. I will provide the identifier of each session so you can easily search them in the session catalog.
Get the data and start predicting the climate change
All data scientists know that the hardest part of any machine learning project is to get good quality data. If you want to start analyzing climate and weather-related data, you are lucky: you can find more than 50 datasets already collected, organized, and published in their Registry of Open Data that anyone can access at no cost.
But how to get started? Learn from Earth scientist Zac Flamig at the chalk talk titled “Methods for analyzing climate and weather data” (WPS202). He promises you to show solving practical problems, like how hot it will be tomorrow, if your building will flood more frequently over the next 100 years, or how much solar power you will generate in the next hour.
So Long, and Thanks for All the Fish
Next time when you eat your tuna nigiri sushi or poke bowl, think a second to the valiant guys who dedicate their carrier to let you eat fish even tomorrow and the day after. Indeed, large-scale fisheries seriously threaten the extinction of tuna species.
Responding to this problem, non-profit organizations like The Nature Conservancy create a sustainable tuna supply chains model. And guess what, they also use machine learning for doing so. Figure out how at the “AI/ML for sustainability innovation: Insight at the edge” breakout session (AIM207) held by Nelson Gonzalez, Head of Global Impact Computing at AWS, and Mark Zimring, Director of Large Scale Fisheries Program at The Nature Conservancy.
Cottage or your arena, build it with zero-carbon emissions
Housing is responsible for one-third of CO2 emissions in the U.S, so constructing sustainable buildings can make a big difference. Amazon wants to tell you the way they do it. As usual, they think in big: they’ve just finished the reconstruction works of their Climate Pledge Arena in Seatle, turning it into the first zero-carbon arena in the world.
Machine learning also has an important role in this field: accurately analyzing energy, water, and air-quality data and building forecasting models can help optimally distribute these resources. Check out their breakout session “Driving sustainable operations at Amazon and Climate Pledge Arena” (AMZ202); maybe you can get some hints on how to build also a cottage in a greener way.
Get the big picture
Our planet is continuously changing. We have many thousands of satellites monitoring it every second, generating hundreds of terabytes of data every day. How to handle this data? And how to notice a relatively small change, for example, a forest fire, or an oil pipeline leakage, as quickly as possible?
You guessed right: using machine learning. Ursa Space Systems does exactly so, and they will tell you how they do it in the chalk talk “Illustrating changes on the earth with insights from space” (WPS209). If you would like to see the big picture, this session is for you.
The bonus track
So, you saved the planet also today. You deserve some chilling: let’s get a fresh handcrafter beer. You would like to try out something new today, but where to get it?
Don’t panic, re:Invent has the response also for this question: find your new beet with the Craft Beer mobile app. Personalized recommendations generated by machine learning models! Real-time feedback taken into account to improve future recommendations! What else could you wish? Drop yourself in the breakout session “Build an app to find your next favorite brew” (BOA302) and get … at least some beer suggestions.
I am Janos Tolgyesi, Machine Learning Solution Architect and ML Team Leader at Neosperience. I am working with ML technologies for five years and with AWS infrastructure for eight years. I love building things, let it be a video analytics application on the edge or a user profiler based on clickstream events. For any questions, you can find me on Medium or Twitter as @jtolgyesi
