Machine learning is a subset of artificial intelligence (AI) which, at the time of writing, has little to do with climate change or our efforts to combat excessive waste. However, many predict that is about to change. Recent developments in machine learning have genuine social value: doctors now use machine learning to make more accurate diagnoses, and self-driving cars are becoming increasingly widespread. Machine learning continues to transform industries and, if we utilize it correctly, it will become a vital tool in our efforts to lead better, more sustainable lives.
The issue, however, is that many of us don’t know what machine learning is and can’t really imagine what a future led by AI might look like. We know what it means to put recycling in the correct bins, and we just have to turn on the news to see the direct impact of excessive waste. But what is machine learning? And how can it help us reduce waste and emissions?
E-Waste and its Effect on Our World
Electronic waste is a disturbing by-product of our rapidly evolving, tech-driven worlds. When an electronic product breaks or is deemed to have outlived its usefulness, it becomes e-waste. Every phone, laptop, household product, or electronic product that we throw away adds to the already staggering pile of e-waste that is held in landfills.
A recent UN study estimated that, in 2019 alone, we threw away 50 million tonnes of e-waste, but only recycled around 10 million metric tonnes. The issue of e-waste is clearly spiraling out of our control and presents a significant threat to the environment and vulnerable human populations worldwide. Simply, we need intelligent solutions that are capable of dealing with the growing issue of e-waste. Many believe those solutions will come from artificial intelligence technologies which are capable of computing and analyzing the massive amounts of data involved in e-waste.
Defining AI + Machine Learning
For those of us less familiar with artificial intelligence and machine learning, it might be helpful to take a step back and provide a few definitions.
Nils J. Nilsson defines AI as “that activity devoted to making machines intelligent, and intelligence is that quality that enables an entity to function appropriately and with foresight in its environment.”
In other words, AI is the use of computing systems to generate intelligence that imitates—and sometimes exceeds—human thought. More specifically, AI’s usefulness lies in the fact that computers are capable of storing and analyzing massive amounts of data more efficiently than humans can. As such, AI is better suited to completing tasks that involve large amounts of data—a human can’t sift through 50 million tonnes of e-waste a year, but a correctly programmed AI certainly can.
Machine learning is a subfield of AI which mimics human learning. Just as we learn things through trial and error, machine learning requires the input of data to learn. The neat thing about machine learning is that it does not require a human at every step: when done correctly, machine learning allows computers to learn and adapt on their own. This is useful as it means computers—which can quickly process vast amounts of data more efficiently than humans—can learn and respond to varied inputs to create solutions that draw from massive volumes of data.
Computer scientist Hilary Mason emphasizes that machine learning’s real value is that “it gives us the ability to learn things about the world from large amounts of data that we as human beings can’t possibly study or appreciate.” Mason also explains that the challenge of machine learning isn’t in simply creating programs but rather in finding ways to effectively apply machine learning to the real world and all its chaotic, unpredictable problems.
As far as unpredictable, chaotic problems go, climate change is something of a final boss. As such, it requires software engineers who specialize in the subfield of machine learning. Machine learning engineers vary slightly from software engineers, as they are required to take on multidisciplinary approaches to data science. Simply, machine learning engineers have to consider the real world that exists beyond the program and must find ways to adapt to meet the demands of society.
Machine Learning’s Vital Role in Reducing E-Waste
Machine learning is already being applied to sustainability projects. Tech giants like Google and IBM are using AI to reduce the overall impact of their operations on the environment, and are now leveraging AI models to advocate for sustainable development.
In the effort to reduce our e-waste and create more sustainable futures, it is clear that we need intelligent AI which is capable of accurately identifying and sorting the 50 million tonnes of e-waste we throw away yearly. As such, there is increasing interest in creating machine-learning programs to tackle e-waste. These developments—in unison with other exciting technologies like augmented reality — will increase the overall efficacy of our efforts to recycle a greater proportion of e-waste.
AI and machine learning can be leveraged on both large and small-scale recycling projects. Companies are already developing waste management AI capable of sorting waste at twice the speed of humans. On a smaller scale, waste bins that contain AI recognition technologies are currently being prototyped internationally, and in Bournemouth, UK, the “Leave Only Footprints” campaign is using drones to gather essential data for waste management AI.
For the time being, reducing our waste is a vital component in fighting climate change and reducing environmental damage. But, with new waste-reduction technologies being trialed and brought into widespread usage, we can feel some hope that the effort to reduce waste and reverse climate change will progress beyond individual responsibility and inefficient recycling bins. Machine learning, in tandem with other high-tech developments, offers us a way to take on the unimaginable scale of e-waste. In the comings years, machine learning will create the deep neural networks of AI knowledge necessary for more efficient recycling schemes—in a strange way, that thought is deeply comforting.