Data is key for modern AI solutions, especially deep learning. Unfortunately, the data-driven nature of deep learning that makes it so powerful when dealing with complex and high-dimensional data, is also at the core of its main weakness: a model is only as good as the data it builds on. In this project, we want to tackle some strong limitations inherent to the standard machine learning paradigm, which makes restrictive assumptions that are problematic in many real-world (“in the wild”) conditions. By addressing these, we want to make a fundamental step towards more powerful deep learning systems that can learn continuously and know how to adapt as new data becomes available, in the context of computer vision. Traditional deep learning relies on the training data being representative for data encountered during system deployment. This is perfect when working with stationary datasets. Yet in practice, data distributions are often non-stationary, i.e., they change over time. This can have a multitude of reasons – think of social trends, seasonal or geographic variations. This calls for a new generation of deep learning methods, able to adapt to new conditions by continuously updating the models based on new training data becoming available. Learning from non-stationary streaming data is, however, still a major challenge requiring fundamental research. In this project, we build on our earlier expertise in continual learning, to realize this ambitious goal. If successful, this will lead to machine learning systems that keep on learning over time, systematically improving their skills and never getting outdated. It also may lower the threshold for applying machine learning, as it reduces the need for a skilled data scientist carefully preparing the data beforehand. As a practical application, we plan to showcase our work’s feasibility, scalability and flexibility in the context of automatic generation of audio descriptions of videos for the visually impaired.

Metal-organic frameworks (MOFs) are crystalline solids with highly regular pores in the nanometer range. The possibility to create a tailored nano-environment inside the MOF pores makes these materials high-potential candidates for integration with microelectronics, e.g. as sensor coatings, solid electrolytes, etc. However, current solvent-based methods for MOF film deposition, a key enabling step in device integration, are incompatible with microelectronics fabrication because of contamination and corrosion issues. VAPORE will open up the path to integrate MOFs in microelectronics by developing a solvent-free chemical vapor deposition (CVD) route for MOF films. MOF-CVD will be the first example of vapor-phase deposition of any type of microporous crystalline network solid and marks an important milestone in processing such materials. Development of the MOF-CVD technology platform will start from a proof-of-concept case and will be supported by the following pillars: (1) Insight in the process, (2) expansion of the materials scope and (3) fine-tuning process control. The potential of MOF-CVD coatings will be illustrated in proof-of-concept sensors. In summary, by growing porous crystalline films from the vapor phase for the first time, VAPORE implements molecular self-assembly as a scalable tool to fabricate highly controlled nanopores. In doing so, the project will enable cross-fertilization between the worlds of nanoscale chemistry and microelectronics, two previously incompatible fields.