Presentation Details
| Benefits and drawbacks of AI in animal behavior: a case study in spider locomotion and communication Erin E Brandt. Grand Valley State University, Allendale, MI, USA |
Abstract
In the fields of animal behavior and biomechanics, there has been a proliferation of machine learning tools to simplify workflows and reduce the time spent on tedious tasks like video tracking. I will discuss benefits and drawbacks of some such tools in the context of my work in spider locomotion and communication. From my perspective at a primarily undergraduate institution, I stress the importance of carefully considering how, when, and why these methods are deployed, in service of developing robust “UI” (undergraduate intelligence) in junior researchers.I argue that some data sets are not well-suited to machine learning tools. In these cases, simpler, less opaque methods may in fact be faster and more accurate, though also more labor-intensive. Importantly, junior researchers working on projects with some “hands on” data processing and analysis gain a deep, immersive knowledge of a dataset, giving them ownership and empowering them to become experts of the dataset. Hands-on immersion develops important scientific skills in junior researchers, such as attention to detail, a critical scientific eye, and the opportunity to take a project further based on their developed insight.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
Content Locked. Log into a registered attendee account to access this presentation.