Workshops: Monday 26 August

On Monday 26 August, we offer a full day (9am-5pm) of workshops that will delve deeper into some of the quantitative tools used in the study of animal behaviour.  Deadline for registrations is 22 July 2019.

You can register for one workshop in the morning and/or one in the afternoon. Please note that places for the "Quantifying behavior with machine-learning workshop (13:00-17:00)" are limited and work on a first come, first serve basis.

Register for Workshops

09:00-12:00 Machine learning in the behavioral sciences

Organizers: Various scientists who are using machine learning at the Max Planck Institute for Animal Behavior (MPI AB)

The principles and ideas of machine learning will be introduced and ongoing studies using this powerful method will be presented.

Number of participants: no limit

13:00-17:00 Quantifying behavior with machine-learning

Organizers: Ian Etheredge and Jake Graving (MPI AB)

More and more, the way we think about behavior is being driven by our ability to measure the postures and movements of individuals. With the increasing availability of high-resolution time series data, how do we turn these into something useful for ethologists? In this workshop we will review the current state of the art for generating postural time-series and how these may be leveraged for describing behavioral repertoires using supervised and unsupervised machine learning.

Experience with Python and Jupyter Notebook is highly recommended, as is an interest in learning more about machine learning.

Number of participants: 30

13:00-16:00 Low-cost automation of behavioral experiments with raspberry pi’s

Organizer: Jolle Jolles (MPI AB)

For the study of animal behavior, standardized testing conditions and optimized procedures play a crucial role in the acquisition of reliable data and appropriate sample sizes for their interpretation. Increasingly tools have become available that help facilitate these important components and automate behavioral experiments. However, getting started with these tools is often challenging. This workshop will focus on the use of low-cost raspberry pi (rpi) computers to improve the study of animal behavior and broaden its potential scope. It is aimed at researchers with little experience in coding or use of raspberry pi’s to help the automation and standardization of behavioral experiments. The workshop will cover the following topics: introduction to rpi and related systems, setting-up a rpi system, standardizing behavioral experiments, camera calibration, automating recordings, data processing, and broader applications.

Number of participants: no limit

13:00-17:00 Recording and analyzing acoustic data

Organizers: Dianne Brunton, Wesley Webb, and Yukio Fukuzama (Massey University); and Jens Koblitz (MPI AB)

In the first part of this workshop we will give an overview of various recording methodologies used to gather data on vocalisations of animals from single receiver devices used in acoustic monitoring to animal-borne sound recording tags to large multi-receiver arrays. Advantages and limitations of the various systems will be outlined. We will run through a case study tutorial that all participants can follow online.In the second part of the workshop, Koe, a web-based software application will be introduced. Koe has two primary functions: classifying acoustic units (which includes visualizing, segmenting, measuring, cataloguing and exporting units), and analyzing syllable differences and sequence structure (syntax). Koe is accessible at koe.io.ac.nz

Requirements: laptop, wifi will be provided

Number of participants: no limit