"Efficient AI for Wildlife Conservation"
Sara M. Beery
Visiting Researcher, Google
Assistant Professor, MIT CSAIL
We require systems to monitor species in real time and in greater detail to quickly understand which conservation and sustainability efforts are most effective and take corrective action. Current ecological monitoring systems generate data far faster than researchers can analyze it, making scaling up impossible without automated data processing. However, ecological data collected in the field presents a number of challenges that current methods, like deep learning, are not designed to tackle. These include strong spatiotemporal correlations, imperfect data quality, fine-grained categories, and long-tailed distributions. Beyond this, many of the areas we seek to monitor are remote, which requires us to work within the constraints of limited bandwidth, power, storage, and computational capacity. I’ll discuss several open challenges in environmental monitoring where more robust, efficient, and adaptable models are needed, and where progress has significant potential for impact.