Publications

Multi-purpose tactile perception based on deep learning in a new tendon-driven optical tactile sensor

Abstract

In this paper, we create a new tendon-connected multi-functional optical tactile sensor, MechTac, for object perception in field of view (TacTip) and location of touching points in the blind area of vision (TacSide). In a multi-point touch task, the information of the TacSide and the TacTip are overlapped to commonly affect the distribution of papillae pins on the TacTip. Since the effects of TacSide are much less obvious to those affected on the TacTip, a perceiving out-of-view neural network (O$^2$VNet) is created to separate the mixed information with unequal affection. To reduce the dependence of the O$^2$VNet on the grayscale information of the image, we create one new binarized convolutional (BConv) layer in front of the backbone of the O$^2$VNet. The O$^2$VNet can not only achieve real-time temporal sequence prediction (34 ms per image), but also attain the average classification accuracy of 99.06%. The experimental results show that the O$^2$VNet can hold high classification accuracy even facing the image contrast changes.

Continue reading

Discovering and visualizing tactics in table tennis games based on subgroup discovery

Abstract

We report on preliminary results to automatically identify efficient tactics of elite players in table tennis games. We define such tactics as subgroups of winning strokes which table tennis experts sought to obtain to train players and adapt their strategy during games. We first report on the creation of such subgroups and their ranking by weighted relative accuracy measure (WRAcc). We then report on representation of the subgroups using visualizations that enabled our expert to provide rapid feedback and hence provided us with guidance towards further improvements of our discoveries

Continue reading

Improving the quality of rule-based GNN explanations

Abstract

Recent works have proposed to explain GNNs using activation rules. Activation rules allow to capture specific configurations in the embedding space of a given layer that is discriminant for the GNN decision. These rules also catch hidden features of input graphs. This requires to associate these rules to representative graphs. In this paper, we propose on the one hand an analysis of heuristic-based algorithms to extract the activation rules, and on the other hand the use of transport-based optimal graph distances to associate each rule with the most specific graph that triggers them.

Continue reading

A Kleene theorem for higher-dimensional automata

Abstract

We prove a Kleene theorem for higher-dimensional automata (HDAs). It states that the languages they recognise are precisely the rational subsumption-closed sets of interval pomsets. The rational operations include a gluing composition, for which we equip pomsets with interfaces. For our proof, we introduce HDAs with interfaces as presheaves over labelled precube categories and use tools inspired by algebraic topology, such as cylinders and (co)fibrations. HDAs are a general model of non-interleaving concurrency, which subsumes many other models in this field. Interval orders are used as models for concurrent or distributed systems where events extend in time. Our tools and techniques may therefore yield templates for Kleene theorems in various models and applications.

Continue reading

A machine learning based approach for the detection of sybil attacks in c-ITS

Abstract

The intrusion detection systems are vital for the sustainability of Cooperative Intelligent Transportation Systems (C-ITS) and the detection of sybil attacks are particularly challenging. In this work, we propose a novel approach for the detection of sybil attacks in C-ITS environments. We provide an evaluation of our approach using extensive simulations that rely on real traces, showing our detection approach?s effectiveness.

Continue reading

Architectures vs the Ports tree: A losing battle?

Abstract

Continue reading

Comparing use-cases of tree-fold vs fold-left, how to fold and color a map

Abstract

In this article we examine some consequences of computation order of two different conceptual implementations of the fold function. We explore a set of performance- and accuracy-based experiments on two implementations of this function. In particular, we contrast the traditional fold-left implementation with another approach we refer to as tree-fold. It is often implicitly supposed that the binary operation in question has constant complexity. We explore several application areas which diverge from that assumption: rational arithmetic, floating-point arithmetic, and Binary Decisions Diagram construction. These are binary operations which degrade in performance as the fold iteration progresses. We show that these types of binary operations are good candidates for tree-fold.

Continue reading

Label-efficient self-supervised speaker verification with information maximization and contrastive learning

Abstract

Continue reading

Using subgroup discovery to relate odor pleasantness and intensity to peripheral nervous system reactions

Abstract

Activation of the autonomic nervous system is a primary characteristic of human hedonic responses to sensory stimuli. For smells, general tendencies of physiological reactions have been described using classical statistics. However, these physiological variations are generally not quantified precisely; each psychophysiological parameter has very often been studied separately and individual variability was not systematically considered. The current study presents an innovative approach based on data mining, whose goal is to extract knowledge from a dataset. This approach uses a subgroup discovery algorithm which allows extraction of rules that apply to as many olfactory stimuli and individuals as possible. These rules are described by intervals on a set of physiological attributes. Results allowed both quantifying how each physiological parameter relates to odor pleasantness and perceived intensity but also describing the participation of each individual to these rules. This approach can be applied to other fields of affective sciences characterized by complex and heterogeneous datasets.

Continue reading

What does my GNN really capture? On exploring internal GNN representations

Abstract

GNNs are efficient for classifying graphs but their internal workings is opaque which limits their field of application. Existing methods for explaining GNN focus on disclosing the relationships between input graphs and the model’s decision. In contrary, the method we propose isolates internal features, hidden in the network layers, which are automatically identified by the GNN to classify graphs. We show that this method makes it possible to know the parts of the input graphs used by GNN with much less bias than the SOTA methods and therefore to provide confidence in the decision process.

Continue reading