-
A benchmark for computational analysis of animal behavior, using animal-borne tags
Authors:
Benjamin Hoffman,
Maddie Cusimano,
Vittorio Baglione,
Daniela Canestrari,
Damien Chevallier,
Dominic L. DeSantis,
Lorène Jeantet,
Monique A. Ladds,
Takuya Maekawa,
Vicente Mata-Silva,
Víctor Moreno-González,
Eva Trapote,
Outi Vainio,
Antti Vehkaoja,
Ken Yoda,
Katherine Zacarian,
Ari Friedlaender
Abstract:
Animal-borne sensors ('bio-loggers') can record a suite of kinematic and environmental data, which can elucidate animal ecophysiology and improve conservation efforts. Machine learning techniques are used for interpreting the large amounts of data recorded by bio-loggers, but there exists no common framework for comparing the different machine learning techniques in this domain. To address this, w…
▽ More
Animal-borne sensors ('bio-loggers') can record a suite of kinematic and environmental data, which can elucidate animal ecophysiology and improve conservation efforts. Machine learning techniques are used for interpreting the large amounts of data recorded by bio-loggers, but there exists no common framework for comparing the different machine learning techniques in this domain. To address this, we present the Bio-logger Ethogram Benchmark (BEBE), a collection of datasets with behavioral annotations, as well as a modeling task and evaluation metrics. BEBE is to date the largest, most taxonomically diverse, publicly available benchmark of this type, and includes 1654 hours of data collected from 149 individuals across nine taxa. In addition, using BEBE, we test a novel self-supervised learning approach to identifying animal behaviors based on bio-logger data, using a deep neural network pre-trained with self-supervision on data collected from human wrist-worn accelerometers. We show that this approach out-performs common alternatives, especially in a setting with a low amount of training data. Datasets, models, and evaluation code are made publicly available at https://github.com/earthspecies/BEBE, to enable community use of BEBE as a point of comparison in methods development.
△ Less
Submitted 10 April, 2024; v1 submitted 18 May, 2023;
originally announced May 2023.
-
Learning interaction rules from multi-animal trajectories via augmented behavioral models
Authors:
Keisuke Fujii,
Naoya Takeishi,
Kazushi Tsutsui,
Emyo Fujioka,
Nozomi Nishiumi,
Ryoya Tanaka,
Mika Fukushiro,
Kaoru Ide,
Hiroyoshi Kohno,
Ken Yoda,
Susumu Takahashi,
Shizuko Hiryu,
Yoshinobu Kawahara
Abstract:
Extracting the interaction rules of biological agents from movement sequences pose challenges in various domains. Granger causality is a practical framework for analyzing the interactions from observed time-series data; however, this framework ignores the structures and assumptions of the generative process in animal behaviors, which may lead to interpretational problems and sometimes erroneous as…
▽ More
Extracting the interaction rules of biological agents from movement sequences pose challenges in various domains. Granger causality is a practical framework for analyzing the interactions from observed time-series data; however, this framework ignores the structures and assumptions of the generative process in animal behaviors, which may lead to interpretational problems and sometimes erroneous assessments of causality. In this paper, we propose a new framework for learning Granger causality from multi-animal trajectories via augmented theory-based behavioral models with interpretable data-driven models. We adopt an approach for augmenting incomplete multi-agent behavioral models described by time-varying dynamical systems with neural networks. For efficient and interpretable learning, our model leverages theory-based architectures separating navigation and motion processes, and the theory-guided regularization for reliable behavioral modeling. This can provide interpretable signs of Granger-causal effects over time, i.e., when specific others cause the approach or separation. In experiments using synthetic datasets, our method achieved better performance than various baselines. We then analyzed multi-animal datasets of mice, flies, birds, and bats, which verified our method and obtained novel biological insights.
△ Less
Submitted 25 October, 2021; v1 submitted 12 July, 2021;
originally announced July 2021.
-
Coherent control of 40-THz optical phonons in diamond using femtosecond optical pulses
Authors:
Tetsuya Kimata,
Kazuma Yoda,
Hana Matsumoto,
Hiroyuki Tanabe,
Fujio Minami,
Yosuke Kayanuma,
Kazutaka G. Nakamura
Abstract:
Coherent control is an optical technique to manipulate quantum states of matter. The coherent control of 40-THz optical phonons in diamond was demonstrated by using a pair of sub-10-fs optical pulses. The optical phonons were detected via transient transmittance using a pump and probe protocol. The optical and phonon interferences were observed in the transient transmittance change and its behavio…
▽ More
Coherent control is an optical technique to manipulate quantum states of matter. The coherent control of 40-THz optical phonons in diamond was demonstrated by using a pair of sub-10-fs optical pulses. The optical phonons were detected via transient transmittance using a pump and probe protocol. The optical and phonon interferences were observed in the transient transmittance change and its behavior was well reproduced by quantum mechanical calculations with a simple model which consists of two electronic levels and shifted harmonic oscillators.
△ Less
Submitted 27 December, 2019;
originally announced December 2019.
-
Improved Activity Forecasting for Generating Trajectories
Authors:
Daisuke Ogawa,
Toru Tamaki,
Tsubasa Hirakawa,
Bisser Raytchev,
Kazufumi Kaneda,
Ken Yoda
Abstract:
An efficient inverse reinforcement learning for generating trajectories is proposed based of 2D and 3D activity forecasting. We modify reward function with $L_p$ norm and propose convolution into value iteration steps, which is called convolutional value iteration. Experimental results with seabird trajectories (43 for training and 10 for test), our method is best in terms of MHD error and perform…
▽ More
An efficient inverse reinforcement learning for generating trajectories is proposed based of 2D and 3D activity forecasting. We modify reward function with $L_p$ norm and propose convolution into value iteration steps, which is called convolutional value iteration. Experimental results with seabird trajectories (43 for training and 10 for test), our method is best in terms of MHD error and performs fastest. Generated trajectories for interpolating missing parts of trajectories look much similar to real seabird trajectories than those by the previous works.
△ Less
Submitted 11 December, 2019;
originally announced December 2019.
-
Multifilament YBa2Cu3O6+x -coated conductors with minimized coupling losses
Authors:
G. A. Levin,
P. N. Barnes,
J. W. Kell,
N. Amemiya,
Z. Jiang,
K. Yoda,
F. Kimura
Abstract:
We report an experimental approach to making multifilament coated conductors with low losses in applied time-varying magnetic field. Previously, the multifilament conductors obtained for that purpose by laser ablation suffered from high coupling losses. Here we report how this problem can be solved. When the substrate metal in the grooves segregating the filaments is exposed to oxygen, it forms…
▽ More
We report an experimental approach to making multifilament coated conductors with low losses in applied time-varying magnetic field. Previously, the multifilament conductors obtained for that purpose by laser ablation suffered from high coupling losses. Here we report how this problem can be solved. When the substrate metal in the grooves segregating the filaments is exposed to oxygen, it forms high resistivity oxides that electrically insulate the stripes from each other and from the substrate. As the result, the coupling loss has become negligible over the entire range of tested parameters (magnetic field amplitudes B and frequencies f) available to us.
△ Less
Submitted 11 May, 2006;
originally announced May 2006.
-
Magnetization Losses in Multiply Connected YBa2Cu3O6+x Coated Conductors
Authors:
G. A. Levin,
P. N. Barnes,
Naoyuki Amemiya,
Satoshi Kasai,
Keiji Yoda,
Zhenan Jiang,
A. Polyanskii
Abstract:
We report the results of a magnetization losses study in experimental multifilament, multiply connected coated superconductors exposed to time-varying magnetic field. In these samples, the superconducting layer is divided into parallel stripes segregated by non-superconducting grooves. In order to facilitate the current sharing between the stripes and thus increase the reliability of the striate…
▽ More
We report the results of a magnetization losses study in experimental multifilament, multiply connected coated superconductors exposed to time-varying magnetic field. In these samples, the superconducting layer is divided into parallel stripes segregated by non-superconducting grooves. In order to facilitate the current sharing between the stripes and thus increase the reliability of the striated conductors, a sparse network of superconducting bridges is superimposed on the striated film. We find that the presence of the bridges does not substantially increase the magnetization losses, both hysteresis and coupling, as long as the number of bridges per length of the sample is not large. These results indicate that it is possible to find a reasonable compromise between the competing requirements of connectivity and loss reduction in an ac-tolerant version of the high temperature coated conductors specifically designed for ac power applications.
△ Less
Submitted 21 October, 2005;
originally announced October 2005.
-
Magnetization Losses in Multifilament Coated Superconductors
Authors:
G. A. Levin,
P. N. Barnes,
N. Amemiya,
S. Kasai,
K. Yoda,
Z. Jiang
Abstract:
We report the results of a study of the magnetization losses in experimental multifilament, as well as control (uniform), coated superconductors exposed to time-varying magnetic field of various frequencies. Both the hysteresis loss, proportional to the sweep rate of the applied magnetic field, and the coupling loss, proportional to the square of the sweep rate, have been observed. A scaling is…
▽ More
We report the results of a study of the magnetization losses in experimental multifilament, as well as control (uniform), coated superconductors exposed to time-varying magnetic field of various frequencies. Both the hysteresis loss, proportional to the sweep rate of the applied magnetic field, and the coupling loss, proportional to the square of the sweep rate, have been observed. A scaling is found that allows us to quantify each of these contributions and extrapolate the results of the experiment beyond the envelope of accessible field amplitude and frequency. The combined loss in the multifilament conductor is reduced by about 90% in comparison with the uniform conductor at full field penetration at sweep rate as high as 3T/s.
△ Less
Submitted 9 September, 2004;
originally announced September 2004.
