Earth Day: Applying AI to Combat Poaching and Assist Conservation

Sustainability and the impact of industrial development on the Earth and its atmosphere is an ever present topic in current news. You’re hard pressed to go onto social media without seeing #sustainability, #zerowaste, #vegan, #climatechange trending, and this can only be a positive step towards saving our planet. Earth Day, however, has been around far longer than today's memes and headlines of global warming, overfarming, and the crisis of endangered species. Back in April of 1970 a huge protest took to the streets of America to stand against the negative impacts of industrial development on the environment. Earth Day is now marked on its anniversary each year and people from all corners of the world come together to petition and march against some of the most damaging impacts that society is having on the world.

Each year, Earth Day focuses on a different aspect of the environment, and this year’s theme is ‘Protect Our Species’. Experts from WWF estimate that the extinction rate is somewhere between 1,000 and 10,000 times higher than the natural extinction rate, and between 0.01 and 0.1% of all species will become extinct each year. Now, this may not sound like a lot, but there are still many unknown species on our planet. To put this into perspective, if there were 100,000,000 different species on Earth, at the extinction rate of just 0.01% we would see at least 10,000 species go extinct every year.  Poaching, deforestation, plastic waste in the ocean, climate change and pollution are killing off species, and these are all caused by human activity.

At the AI for Good Summit in San Francisco this June 20 - 21 we will be covering topics such as sustainability and the environment, and Bistra Dilkina, Assistant Professor at USC will be presenting her work on using machine learning to predict where poachers may strike for the protection of wildlife.

Poaching is a global issue. Different groups of people from all corners of the globe are threatening the lives of countless species. Taking into consideration the amount of different species, the movements of the poachers, and their different tactics immediately introduces millions of data points. Attempting to draw conclusions and make predictions from this data is challenging for humans. Many species are being specifically targeted by poachers which is leading to the threat of extinction, and this is not just limited to animals in the wild. Many conservation groups are targeted by poachers and are looking for ways to protect their animals. Bistra is working to develop mathematical optimised AI to help trace the patterns of poachers and help predict when they may be most likely to strike, allowing conservationists to be prepared and protect their animals. She is working with several PhD students to generate algorithms that analyze this data about poachers taking into consideration their hotspots and activities. With this data,  they then create an algorithm to map the optimum patrol route. This helps to identify the best routes to intercept poachers and to avoid traps and snares. The model they have created is called PAWS (Protection Assistant for Wildlife Security) and is in operation in nearly 600 wildlife parks, where Bistra believes that AI can help save endangered species and their habitats.

Bistra first became interested in conservation whilst working on her PhD at Cornell where her advisor prompted the thought of harnessing AI for a sustainable future. In an interview with USc, Bistra explained that ‘managing the biodiversity extinction crisis requires wise decision-making processes able to account for the limited resources available. In other words, conservation efforts need to make a big impact on a small budget. You have to make sure that the investment and the method is as effective as possible—that’s where our algorithms come in.’ NGOs have huge amounts of data that they’re keen to see put to use, so Bistra and her team are using this to try and solve these real world problems. There are many challenges such as multiple species needing conservation in the same area swaying budget priorities, but the algorithms help to take this into account.

Bistra is working with the next generation of experts in conservation and AI, and she said that ‘this semester [she is] teaching an undergraduate class on AI for social good—the students are so hungry to learn about AI and they are so excited about how it can make a difference in the world.’

Whilst it’s still early days for PAWS, it’s already having a huge impact; ‘The predictive model guided them to places they wouldn’t normally go and they confiscated over 500 snares in one month. By comparison, they found only 101 snares per month on average in the rest of 2018! We are hoping that rangers around the world will soon be able to use PAWS.’

At the AI for Good Summit we will also hear from Pushpendra Rana, Postdoctoral Researcher at University of Illinois who will give an overview of the impact of machine learning for environmental sustainability. He will explain the importance of these techniques and share how they are revolutionizing how we understand and design effective strategies to deal with forest and other natural resource sustainability challenges.

Additional confirmed speakers include Carlos Felipe Gaitan Ospina, Chief Climate Scientist & Head of AI at ClimateAI; Priya Vijayarajendran, CTO Vice President Applied AI at IBM; George Kantor, Founder at Farmview and more.