Some of the biggest names in AI research have laid out a roadmap for how machine learning can help save our planet and humanity from imminent peril.
The report covers possible machine-learning (ML) interventions across 13 domains from electricity systems to farms and forests to climate prediction. Within each domain, it breaks out the contributions for various sub-disciplines within machine learning, including computer vision, natural language processing, and reinforcement learning.
Some of the recommendations are also categorized into three buckets: “high leverage” for problems well-suited to ML that will likely be fruitful and impactful; “long-term” for solutions that won’t have payoffs until 2040; and “high risk” for pursuits that have less certain outcomes, either because the technology isn’t mature or not enough is known to assess the consequences. Many of the recommendations also summarize existing efforts that are already happening but not yet at scale.
The report’s compilation was led by David Rolnick, a postdoctoral fellow at the University of Pennsylvania, and advised by several high-profile figures, including Andrew Ng, the co-founder of Google Brain and a leading AI entrepreneur and educator; Demis Hassabis, the founder and CEO of DeepMind; Jennifer Chayes, the managing director of Microsoft Research; and Yoshua Bengio, who recently won the Turing Award for his contributions to the field. While the researchers offer a very comprehensive list of some of the major areas where machine learning can contribute, they also note it is not a silver bullet. Ultimately, policy will be the main driver for effective large-scale climate action.
Here are just ten of the “high leverage” recommendations from the report. Read the full version of it here.
1. Improve predictions of how much electricity we need. If we’re going to rely on more renewable energy sources, utilities will need better ways of predicting how much energy is needed, in real time and over the long term. ML algorithms already exist that can forecast energy demand, but they could be improved to reach far higher granularity by taking into account finer local weather and climate patterns or different households’ behavioral patterns. Efforts to make the algorithms more explainable could also help utility operators to better interpret their outputs and schedule when to bring renewable sources online.
2. Discover new materials. Scientists need to develop new materials that store and harvest energy more efficiently. But the process of discovering new materials is typically slow and imprecise. ML techniques can accelerate things by finding, designing, and evaluating new chemical structures with the desired properties, such as to create solar fuels, which can store energy from sunlight. This logic also extends to finding more efficient carbon dioxide absorbents or structural materials that use a lot less carbon to create. The latter could one day replace steel and cement—the production of which accounts for nearly 10% of all global greenhouse gas emissions.
3. Optimize how freight is routed. Shipping goods around the world is a complex and often highly inefficient process that involves the interplay of different shipment sizes, different types of transportation, and a changing web of origins and destinations. An ML system could improve the scheduling and routing of freight operations to bundle together as many shipments as possible and minimize the total number of trips. It would also be more resilient to transportation disruptions.
4. Lower barriers to electric vehicle adoption. Electric vehicles are a key strategy for decarbonizing transportation and face several adoption challenges that are suitable for ML. Algorithms can improve battery energy management to increase the mileage of each charge and reduce range anxiety, for example. They can also model and predict aggregate charging behavior to help electric grid operators meet and manage their load.
5. Help make buildings more efficient. ML-based intelligent control systems can dramatically reduce a building’s energy consumption by taking weather forecasts, building occupancy, and other environmental conditions into account to adjust the heating, cooling, ventilation, and lighting needs in an indoor space. A smart building could also talk directly with the grid to reduce how much power it is using if there’s a scarcity of low-carbon electricity supply at any given time.
6. Create better estimates of how much energy we are consuming. Many regions of the world have little to no data on their energy consumption and greenhouse gas emissions, which can be a major obstacle for designing and implementing effective mitigation strategies. Computer vision techniques can extract building footprints and characteristics from satellite imagery to feed ML algorithms that can estimate city-level energy consumption. The same techniques could also identify which buildings should be retrofitted to maximize their efficiency.
7. Optimize supply chains. In the same way that ML can optimize shipping routes, it can also minimize the inefficiencies and emissions of the rest of the supply chain across the food, fashion, and consumer goods industries. Better predictions of supply and demand will significantly reduce production and transportation waste, while targeted recommendations for low-carbon product alternatives could shift buyer behavior toward more environmentally-friendly consumption.
8. Make precision agriculture possible at scale. Much of modern-day agriculture is dominated by monoculture, the practice of producing a single crop on a large swath of farmland. This approach makes it easier for farmers to manage their fields with tractors and other basic automated tools, but it also strips the soil of its nutrients and reduces its productivity. As a result, many farmers rely heavily on nitrogen-based fertilizers, which can convert into nitrous oxide, a greenhouse gas 300 times more potent than CO2. Robots run on ML software could help farmers manage a mix of crops more effectively at scale, while algorithms could help farmers predict what crops to plant when, regenerating the health of their land and reducing the need for fertilizers.
9. Improve deforestation tracking. Deforestation contributes to roughly 10% of global greenhouse gas emissions but tracking and preventing it is usually a tedious manual process that takes place on the ground. Satellite imagery and computer vision techniques can automatically analyze the loss of tree cover at a much greater scale, and sensors on the ground combined with ML algorithms for detecting chainsaw sounds can help local law enforcement officials stop illegal activity.
10. Nudge consumers to change how we shop. Just as advertisers have successfully used ML to segment and target consumers, the same techniques can be used to help us behave in more environmentally-aware ways. Different consumers with different preferences could receive tailored interventions to promote their enrollment in energy saving programs, for example, or close their psychological distance to climate change so they are more likely to support productive climate policies.