In recent years, novel technologies based on quantum communication such as quantum key distribution have made the transition from theory to reality. In this project, we will consider another important cryptographic task, namely quantum position verification. Quantum position verification is the principal cryptographic primitive for quantum position-based cryptography. Here, the idea is to use a party's position instead of, for example, a secret key as credential. Possible applications include increasing the security of online banking. It has been proven that secure position verification is impossible without quantum resources. The aim of the project is to remove the last remaining obstacles towards a first experimental realization of quantum position verification. The project's aims are three-fold: First, we will solve the practical challenges standing in the way of realizing the measuring protocol which is one of the first protocols proposed for quantum position verification (and for which the scientific coordinator is one of the co-authors of its security proof). The measuring protocol allows to do secure position verification with only a single qubit and is reminiscent of the BB84 protocol, but where the choice of basis is dictated by evaluating a classical function. Second, we investigate the entanglement cost of attackers in order to reduce the exponential gap between the best theoretical security guarantees and the best known attacks. Finally, we will study how quantum position-verification can be integrated into more complex protocols, considering for example device-independence and composability of protocols for position verification. Thus, this project will contribute to making quantum position-based cryptography a reality.

A treebank is a collection of sentences annotated with their syntactic structures (dependency tree or constituency tree). Such resources are important for data-driven linguistic studies. However, constructing treebanks is particularly expensive and requires expert linguistic knowledge, which makes automatique methods based on syntactic parsing (the task of predicting a syntactic tree for a given input sentence) and machine learning very appealing. Furthermore, though many treebanks are available for multiple domains of written French, large-scale treebanks for spontaneous spoken French are scarce. This project deals with syntactic parsing of spontaneous spoken French. Current research in spoken language parsing mostly use gold-standard transcriptions as input, which does not make them robust to automatic speech recognition (ASR) errors. Moreover, transcriptions are an abstraction of speech where prosodic information are no longer accessibles. We propose to investigate approaches that would allow us to perform jointly ASR and syntactic parsing of speech, with a double aim: (i) allow the parser to use prosodic information in the speech signal, (ii) allow ASR to use syntactic information to improve disambiguation quality. To do so, we will design new architectures for parsing speech directly from the speech signal based on: - cross-modal learning methods, in order to leverage partially annotated (spoken or written) data - the fusion of acoustic (wav2vec) or linguistic (BERT/FlauBERT) pretrained representations. We plan to evaluate our proposals on available spoken French treebanks (such as: Orféo, Rhapsodie, paris-stories, ODIL-syntax). Morover, we will carry the extrinsic evaluation of our models to assess to what extent syntax-enhanced speech encoders lead to improvements in downstream speech processing tasks, such as speech translation or spoken language understanding (SLU).

Greatly raising the bar from previous generation upgrades, fifth-generation (5G) mobile systems are promising fiber-like connectivity and a “faster-than-thought” Internet experience to a diverse ecosystem of users and devices with highly variable attributes and demands. As a result, the successful deployment of 5G systems requires a drastic rethinking of existing resource allocation methodologies with the goal of enabling resource-limited users and devices to adapt “on the fly” to a rapidly – and unpredictably – changing wireless landscape. So far, the vast majority of works on wireless resource allocation (spectrum, power, etc.) has focused on two limit cases: In the static regime, the attributes of the network are assumed effectively static and the system’s optimality analysis relies on techniques from optimization, game theory and (optimal) control. On the other hand, in the so-called stochastic regime, the network is assumed to evolve randomly following some stationary probability law, and the allocation of wireless resources is optimized using tools from stochastic optimization and control. In practical wireless networks however, both assumptions fail because of factors that introduce an unpredictable variability to the system (such as user mobility, users going arbitrarily on- and off-line, non-random channel fluctuations, etc.). As a result, existing resource allocation schemes do not – in fact, cannot – apply in this setting because “optimum” target states no longer exist, either static or in the mean. In view of the above, ORACLESS envisions a drastic turn towards a flexible, oracle-less resource allocation paradigm (i.e. without any prior system knowledge) based on the deployment of online optimization protocols at the network’s edge (the system’s wireless devices). Our targeted breakthrough will thus be to develop highly adaptive resource allocation schemes that are provably capable of tracking unpredictable changes in the network: in so doing, the distribution of online optimization protocols at the device end will act as an effective multiplier of wireless resources, ensuring at the same time the system’s robust, self-healing operation in the presence of variabilities and fluctuations. Specifically, driven by the leading role played by massive multiple-input, multiple-output (MIMO) and cognitive radio (CR) technologies in the ongoing transition to 5G mobile systems, we intend to focus on the following objectives of high practical relevance: 1. Adaptive optimization schemes for massive MIMO systems: in particular, our aim will be to develop adaptive algorithms for tracking optimum transmit attributes in unpredictable MIMO environments, to safeguard the algorithms’ efficient operation under feedback imperfections, and to resolve the semidefinite computational bottlenecks that arise in the case of massively large antenna arrays. 2. Online policies for dynamic and opportunistic spectrum access: namely, methods for following the equilibrium state of a dynamic network with CR capabilities, to ensure the methods’ robustness against asynchronicities, and to mitigate the effects of user heterogeneity on their performance. Tackling these objectives will require an interdisciplinary blend of techniques from online optimization, learning, game theory, stochastic approximation, and information theory. Thus, given the diverse expertise of the project’s members, the ORACLESS team is uniquely poised to successfully address the challenges identified above and, in so doing, to establish a solid presence in the mature stage of the 5G standardization process where the real identity of 5G will be uncovered.

The PhyFlex project addresses the lack of flexibility in the control of Tangible User Interaction (TUI). On the contrary to previous work, this project addresses a new research question: universal and flexible physical control. PhyFlex opens new perspectives on TUI based on the flexibility provided by the dynamic change of shapes of input devices. Shape-Changing TUI consist of physical User Interfaces (UI) whose shape can be changed by the user or the system. For instance, a tangible knob can be reshaped continuously to increase its resolution when its diameter increases. Providing the physical flexibility of control that is lacking in TUIs could reshape the User Interfaces of tomorrow. In particular, in the critical domains where TUIs are widely and successfully used, the lack of flexibility costs money, usability, mobility and, sometimes, even security. As a consequence, the problem addressed by the project is a very important one. However, to provide physical flexibility to TUIs, the project will need to overcome several obstacles: (1) current knowledge lacks the fundamentals of interaction with physically flexible UI, (2) current HCI theories are inconsistent with physically flexible UI, (3) the design of efficient interaction techniques with physically flexible controls is an unknown area of HCI, (4) material and hardware technologies to build prototypes are not mature yet, and (5) current way of implementing interaction in software are inappropriate. The intended breakthrough of this project is to bring the flexibility of control to physical user interfaces through physical shape-change. Toward this goal, the project will devise a new theory for physically flexible User Interfaces. Exploratory User studies will unveil the opportunities and limits of physical shape-change to support flexibility of control. Interaction techniques will explore devices that change in resolution, definition, number of dimensions and range. Prototypes combining physical, hardware and software will be built. User studies demonstrating the benefits and drawbacks of prototypes will be conducted to feed the theory. Tools to build such user interfaces have yet to be invented, for the research community to take up the results as well as for transfer. The originality of our approach is fourfold: (1) We want to address the fundamentals of interaction instead of taking a technologically-driven approach, (2) We want to address the continuous flexibility of control through shape-change, (3) We want to address the challenges of computer science, (4) We want to ensure retro-compatibility with current interaction. Future applications are numerous, ranging from tangible controls for interactive visualization, mixing console, control of LED lighting, cockpit or camera control, etc. to personal computers. To achieve the objectives, the project brings together a CNRS researcher, coordinator of the project, specialist in Human-Computer Interaction with mixed physical-digital UI, as well as a team of nine young researchers (Researchers, Assistant professors, Master and PhD students and an engineer).

In a broad range of application areas, data is produced at a scale so that it cannot be managed with conventional technologies. This has led to the emergence of the Big Data phenomenon requiring new tools for collecting, querying and mining large sets of data. Heterogeneous data collected in a massive way or broadcast, must indeed be cleaned, crossed and enriched, filtered and aggregated to form in fine products rich in semantics and strategic for analysis and decision making. We move from a huge amount of data to oceans of knowledge with at the heart of this transition, new scientific and technological advances that bring innovations in socio-economic and scientific world. Providing support to the continuum "data-information-knowledge-making" requires to: • manage, organize, access to masses of data from many different sources (Volume), with large differences in terms of content, structure and semantics (Variety), with a high rate of change (Velocity) and whose quality is more or less guaranteed / certified (Veracity) • extract relevant knowledge and some added value using data analysis and data mining processes. The presence of inaccuracies, inconsistencies, errors, expressions of opinions, etc. makes knowledge discovery and decision making complex tasks. Scientific communities concerned with these challenges are those of data management and databases, information retrieval, statistics and data analysis, knowledge discovery from data, machine learning, artificial intelligence, or visualization. The general objective of the MDK is to provide a framework that enables exchange of ideas and drives collective efforts from all these communities, on research and innovation in the field of Big Data and Knowledge analysis and management. The goal is to provide prospective studies, to make recommendations and to propose concrete actions. This collective interdisciplinary effort will be accompanied by a strategy at national and European level.