Search

Written reflective practice is a regular exercise pre-service teachers perform during their higher education. Usually, their lecturers are expected to provide individual feedback, which can be a challenging task to perform on a regular basis. In this paper, we present the first open-source automated feedback tool based on didactic theory and implemented as a hybrid AI system. We describe the components and discuss the advantages and disadvantages of our system compared to the state-of-art generative large language models. The main objective of our work is to enable better learning outcomes for students and to complement the teaching activities of lecturers., Comment: 2 figures

Saliency methods attempt to explain deep neural networks by highlighting the most salient features of a sample. Some widely used methods are based on a theoretical framework called Deep Taylor Decomposition (DTD), which formalizes the recursive application of the Taylor Theorem to the network's layers. However, recent work has found these methods to be independent of the network's deeper layers and appear to respond only to lower-level image structure. Here, we investigate the DTD theory to better understand this perplexing behavior and found that the Deep Taylor Decomposition is equivalent to the basic gradient$\times$input method when the Taylor root points (an important parameter of the algorithm chosen by the user) are locally constant. If the root points are locally input-dependent, then one can justify any explanation. In this case, the theory is under-constrained. In an empirical evaluation, we find that DTD roots do not lie in the same linear regions as the input - contrary to a fundamental assumption of the Taylor theorem. The theoretical foundations of DTD were cited as a source of reliability for the explanations. However, our findings urge caution in making such claims., Comment: Published in Transactions on Machine Learning Research (TMLR)

Understanding the emergence of collective behaviour has long been a key research focus in the natural sciences. Besides the fundamental role of social interaction rules, a combination of theoretical and empirical work indicates individual speed may be a key process that drives the collective behaviour of animal groups. Socially induced changes in speed by interacting animals make it difficult to isolate the effects of individual speed on group-level behaviours. Here, we tackled this issue by pairing guppies with a biomimetic robot. We used a closed-loop tracking and feedback system to let a robotic fish naturally interact with a live partner in real time, and programmed it to strongly copy and follow its partner's movements while lacking any preferred movement speed or directionality of its own. We show that individual differences in guppies' movement speed were highly repeatable and shaped key collective patterns: higher individual speeds resulted in stronger leadership, lower cohesion, higher alignment and better temporal coordination in the pairs. By combining the strengths of individual-based models and observational work with state-of-the-art robotics, we provide novel evidence that individual speed is a key, fundamental process in the emergence of collective behaviour.

Attribution methods aim to explain a neural network's prediction by highlighting the most relevant image areas. A popular approach is to backpropagate (BP) a custom relevance score using modified rules, rather than the gradient. We analyze an extensive set of modified BP methods: Deep Taylor Decomposition, Layer-wise Relevance Propagation (LRP), Excitation BP, PatternAttribution, DeepLIFT, Deconv, RectGrad, and Guided BP. We find empirically that the explanations of all mentioned methods, except for DeepLIFT, are independent of the parameters of later layers. We provide theoretical insights for this surprising behavior and also analyze why DeepLIFT does not suffer from this limitation. Empirically, we measure how information of later layers is ignored by using our new metric, cosine similarity convergence (CSC). The paper provides a framework to assess the faithfulness of new and existing modified BP methods theoretically and empirically. For code see: https://github.com/berleon/when-explanations-lie, Comment: Published in ICML 2020, Camera Ready Version

The study of animal behavior increasingly relies on (semi-) automatic methods for the extraction of relevant behavioral features from video or picture data. To date, several specialized software products exist to detect and track animals' positions in simple (laboratory) environments. Tracking animals in their natural environments, however, often requires substantial customization of the image processing algorithms to the problem-specific image characteristics. Here we introduce BioTracker, an open-source computer vision framework, that provides programmers with core functionalities that are essential parts of a tracking software, such as video I/O, graphics overlays and mouse and keyboard interfaces. BioTracker additionally provides a number of different tracking algorithms suitable for a variety of image recording conditions. The main feature of BioTracker is however the straightforward implementation of new problem-specific tracking modules and vision algorithms that can build upon BioTracker's core functionalities. With this open-source framework the scientific community can accelerate their research and focus on the development of new vision algorithms., Comment: 4 pages, 2 figures

Responding towards the actions of others is one of the most important behavioural traits whenever animals of the same species interact. Mutual influences among interacting individuals may modulate the social responsiveness seen and thus make it often difficult to study the level and individual variation in responsiveness. Here, open-loop biomimetic robots that provide standardized, non-interactive social cues can be a useful tool. These robots are not affected by the live animal's actions but are assumed to still represent valuable and biologically relevant social cues. As this assumption is crucial for the use of biomimetic robots in behavioural studies, we hypothesized (i) that meaningful social interactions can be assumed if live animals maintain individual differences in responsiveness when interacting with both a biomimetic robot and a live partner. Furthermore, to study the level of individual variation in social responsiveness, we hypothesized (ii) that individual differences should be maintained over the course of multiple tests with the robot. We investigated the response of live guppies (Poecilia reticulata) when allowed to interact either with a biomimetic open-loop-controlled fish robot—‘Robofish’—or with a live companion. Furthermore, we investigated the responses of live guppies when tested three times with Robofish. We found that responses of live guppies towards Robofish were weaker compared with those of a live companion, most likely as a result of the non-interactive open-loop behaviour of Robofish. Guppies, however, were consistent in their individual responses between a live companion and Robofish, and similar individual differences in response towards Robofish were maintained over repeated testing even though habituation to the test environment was detectable. Biomimetic robots like Robofish are therefore a useful tool for the study of social responsiveness in guppies and possibly other small fish species.

Responding towards the actions of others is one of the most important behavioural traits whenever animals of the same species interact. Mutual influences among interacting individuals may modulate the social responsiveness seen and thus make it often difficult to study the level and individual variation in responsiveness. Here, open-loop biomimetic robots that provide standardized, non-interactive social cues can be a useful tool. These robots are not affected by the live animal's actions but are assumed to still represent valuable and biologically relevant social cues. As this assumption is crucial for the use of biomimetic robots in behavioural studies, we hypothesized (i) that meaningful social interactions can be assumed if live animals maintain individual differences in responsiveness when interacting with both a biomimetic robot and a live partner. Furthermore, to study the level of individual variation in social responsiveness, we hypothesized (ii) that individual differences should be maintained over the course of multiple tests with the robot. We investigated the response of live guppies (Poecilia reticulata) when allowed to interact either with a biomimetic open-loop-controlled fish robot—‘Robofish’—or with a live companion. Furthermore, we investigated the responses of live guppies when tested three times with Robofish. We found that responses of live guppies towards Robofish were weaker compared with those of a live companion, most likely as a result of the non-interactive open-loop behaviour of Robofish. Guppies, however, were consistent in their individual responses between a live companion and Robofish, and similar individual differences in response towards Robofish were maintained over repeated testing even though habituation to the test environment was detectable. Biomimetic robots like Robofish are therefore a useful tool for the study of social responsiveness in guppies and possibly other small fish species.

The honeybee is a fascinating model animal to investigate how collective behavior emerges from (inter-)actions of thousands of individuals. Bees may acquire unique memories throughout their lives. These experiences affect social interactions even over large time frames. Tracking and identifying all bees in the colony over their lifetimes therefore may likely shed light on the interplay of individual differences and colony behavior. This paper proposes a software pipeline based on two deep convolutional neural networks for the localization and decoding of custom binary markers that honeybees carry from their first to the last day in their life. We show that this approach outperforms similar systems proposed in recent literature. By opening this software for the public, we hope that the resulting datasets will help advancing the understanding of honeybee collective intelligence.