Animal Behavior Challenge (ABC2018) for understanding animal behavior

Organized by tttamaki - Current server time: Dec. 15, 2018, 4:57 a.m. UTC

First phase

Development Phase
March 1, 2018, midnight UTC

End

Competition Ends
Aug. 1, 2018, 11:59 p.m. UTC

Animal Behavior Challenge: ABC2018

Predicting gender of bird from its GPS trajectory: Sponsored by Technosmart

Researchers attach GPS loggers on birds' body for collecting trajectories of birds to understand how they navigate themselves, even across a big ocean.

What is the cue and how they use it ? If male and female birds could use different cues, they must have different trajectories.

Can we find the difference of male/female trajectories and discover a new science knowledge ?

This is the task of this competition.

Data collection

Go to the colony in a small island in Japan

Fieldwork was performed on the population of streaked shearwaters breeding on Awashima Island (38°28′N, 139°14′E), Japan, in a colony that contains approximately 84000 shearwaters.

Find the birds

Parents of this species, Streaked shearwaters (Calonectris leucomelas), attend their nests and feed chicks during the breeding season. The shearwaters feed on fish such as Japanese anchovy in areas up to 1000 km from the breeding colony.

Attach the small logger with a great care

Data loggers (Technosmart Europe) are attached to the leg. Note that the procedures used in the field study for collecting the data were approved by the Animal Experimental Committee of Nagoya University.

Release, then catch the bird with the logger on its leg

Some days after releasing birds, the same birds should be found around the next so that the loggers are retrieved and the data are collected. Note that the trajectories used in this competition have been preprocessed by data cleaning for rejecting incorrect GPS data.

Examples of male and female trajectories

Blue trajelctory is male, red one is female.

About us

Systems Science of Bio-Navigation

Grant-in-Aid for Scientific Research on Innovative Areas from 2016 to 2021

Navigation is a fundamental behavior of animals including human. In navigation, the following three functions are required: the acquisition of dynamically-changing information from external and internal environment, the choice of route and destination based on the information, and the behavioral regulation to reach the destination. We aim for systems science of bio-navigation to understand the “algorithms” for the navigation of animals. To this end, we bring together experts from control engineering, data science, animal ecology, and neuroscience, and jointly work on how to measure, analyze, understand, and verify bio-navigation.

By hosting competitions "Animal Behavior Challenge (ABC)", we hope to boost our understanding the algorithms for animal navigation.

Please visit our website.

Sponsor

This challenge is sponsored by Technosmart (see Prize page). All trajectory data used in this challenge were obtained by GPS loggers of Technosmart.

Technosmart

Rules

You may submit 5 submissions every day and 50 in total in the development phase.

Accuracy

Submitted results are evaluated by the accuracy, widely used for binary classification problems.

  • acc = (number of correctly classified test trajectories) / (total number of test trajectories)

It is 0 if all predictions are wrong, while 1 if perfect.

Submission Format of the result

There are 275 test trajectory, and classification results should be a 275-line text file, which contains 0 (male) or 1 (female) for each of test trajectory.

Each line has the label of the corresponding test trajectory; that is, line 0 is the label of the test trajectory 000.csv.

Predicted label is binary (character):

  • male: 0
  • female: 1

Here is an example:
=================
1
1
1
0
1
0
0
...
=================

  • The name of this file should be "y_submission.txt" in the zip file to be uploaded.
  • You need to zip the file to submit. (Do not put it in a folder in the zip file, do not include any folders. The file must be in the top in the zip file. ) Therefore the uploaded filename maybe "y_submission.txt.zip", or any zip file of the following contents:
    • anyfilename.zip/
      • y_submission.txt
    • On linux or mac, the command "zip y_submission.txt.zip y_submission.txt" is enough.

 

Task

Classifying GPS trajectories of birds into male or female

Trajectory file format

A single CSV file (000.csv, 001.csv, ...) contains a trajectory of a trip,
and each line represents the information of a GPS location of a shearwater.
In addition to longitude and latitude, some other information is provided;
elapsed time and local clock time, solar azimuth and elevation angles.

  • float: longitude
  • float: latitude
  • float: sun azimuth [degree] clockwise from the North
  • float: sun elevation [degree] upward from the horizon
  • int: (1) daytime (between sunrise and sunset), or (0) nighttime
  • int: elapsed time [second] after starting the trip
  • clock: local time (hh:mm:ss)
  • int: days (starts from 0, and increments by 1 when the local time passes 23:59:59)

Float values are of the format %.5f, and fields are separated by a single comma.

Here is an example:

=================
139.29220,38.56632,76.42170,-4.45122,0,0,04:54:03,0
139.29300,38.56763,76.58196,-4.25726,0,60,04:55:03,0
139.29400,38.57053,76.73674,-4.06880,0,118,04:56:01,0
139.29620,38.57563,76.89729,-3.87201,0,178,04:57:01,0
...
=================

  • Different trajectories have different number of GPS locations.
  • The time interval between successive two GPS locations is approximately one minute (60 seconds) when GPS works well, otherwise interval may vary from one to several minutes, even hours and days.
  • Trajectories in the training and test sets are in the same format.
  • Ground truth labels for the training set are given in a separate file.

 

Labels: gender, or male/female

A single txt file of ground truth labels of the training set is provided.
Each line has the label of the corresponding training trajectory; that is, line 0 is the label of the training trajectory file 000.csv.

Label is binary (character):

  • male: 0
  • female: 1

Here is an example:
=================
1
1
1
0
1
0
0
...
=================

 

Stats: Numbers of the dataset

  • Training set
    • 326 male trajectories
    • 305 female trajectories
    • 631 in total
  • Test set
    • 275 trajectories

Prizes

Prize sponsort by Technosmart

Thanks to Tehcnosmart, the sponosor of this challenge, winners will recieve the following prizes;

  • First place: 500EUR
  • Second place: 300EUR
  • Third place: 200EUR

Note that, to recieve the prizes, winners must publish their methods and/or codes available online (for example, as a technical report on arXiv and github) upon request by the organizer.

Travel grant

In addition, the first winner will recieve

  • Flight/surface ticket (+ hotel for a few days) to Kyoto, one of the most beautiful cities in Japan especially in autumn, for attending the award seremony and presenting the method in the poster session of our symposium.

Note that

  • In case of team submission, the travel grant goes to one person only who is a representative.
  • The prize will go to the next winner if the first winner withdraws the attendance and the poster presentation for any reaasons (e.g., personal or VISA issues).
  • We will provide an invitation letter and the winner have to obtain a VISA by himself/herself.

Terms and Conditions

This challenge is governed by the general ChaLearn contest rules.

Results and code

  1. First place 72.00% by fayzur20
  2. Second place 70.18% by rook_s_pawn
  3. Third place 69.09% by tanaka and pankajvshrma

Development Phase

Start: March 1, 2018, midnight

Description: Develop your models and submit prediction results. Scores are shown with randomness (10% of your submission labels are randomly modified for avoiding cheat).

Final Phase

Start: Aug. 1, 2018, midnight

Description: Final phase (no submission, your submissions from the previous phase are forwarded. No randomness is used for evaluation).

Competition Ends

Aug. 1, 2018, 11:59 p.m.

You must be logged in to participate in competitions.

Sign In