Binz et al. highlight the potential of meta-learning to greatly enhance the flexibility of AI algorithms, as well as to approximate human behavior more accurately than traditional learning methods. We wish to emphasize a basic problem that lies underneath these two objectives, and in turn suggest another perspective of the required notion of “meta” in meta-learning: knowing what to learn.

Companies such as Toyota, Samsung, Alibaba, and Lenovo are already innovating with start-ups in the deep-tech field—a group of emerging technologies based on scientific discoveries or meaningful engineering innovations, offering a substantial advance over established technologies, and seeking to tackle some of the world’s fundamental challenges. Currently, this group usually encompasses artificial intelligence, advanced materials, biotechnology, blockchain, robotics and drones, photonics and electronics, and quantum computing. This phenomenon is increasing at speed. In past years, deep-tech startups have received increased attention (...) among corporations, entrepreneurs, investors, and media. Investment in deep-tech start-ups has more than quadrupled over a five-year period, from $15 billion in 2016 to more than $60 billion in 2020, approximately. The average disclosed amount per private investment event for these start-ups and scale-ups has grown 3.4 times between 2016 and 2020. In East and Southeast Asia, some of the corporate venturing activity is concentrated in nine regions: mainland China, Hong Kong, Indonesia, Japan, South Korea, Singapore, Thailand, Taiwan, and Vietnam. On average, this region has a higher adoption rate of corporate venturing (57%), compared to that of Latin American countries (40%), while having room for growth compared to the adoption rate in other regions such as the United States (90%). (shrink)

Equipped with the ultimate query answering system, computers would finally be in a position to address all our information needs in a natural way. In this paper, we describe how Language and Computing nv (L&C), a developer of ontology-based natural language understanding systems for the healthcare domain, is working towards the ultimate Question Answering (QA) System for healthcare workers. L&C’s company strategy in this area is to design in a step-by-step fashion the essential components of such a system, each component (...) being designed to solve some one part of the total problem and at the same time reflect well-defined needs on the prat of our customers. We compare our strategy with the research roadmap proposed by the Question Answering Committee of the National Institute of Standards and Technology (NIST), paying special attention to the role of ontology. (shrink)

Review of Mercè Prats, Une parole attendue. La circulation des polycopiés de Teilhard de Chardin, préface d’Étienne Fouilloux, Paris, Salvator, 2022, 252 p., 21 €.

The article seeks to elucidate the status of transcendence in the historiography of secularization through the perspective of collective memory. It discusses two typological models dealing with the basic metaphysical problem concerned with the presence and meaning of transcendence in real human existence. According to the first, the historical reality of secularization causes a break from the collective memory whose roots are in religion. In contrast, the second model considers that despite the deep transformations in the status of religion in (...) a reality of secularization, an experience of historical continuity may also occur there. These models denote the two poles in the argument about the meaning and value of history for modern people. The article suggests a phenomenological analysis of the two models and criticizes their deficiencies. Finally, the “tension model” is outlined as a third alternative that aims at overcoming the binary situation created by the first two in favor of a perspective that necessitates and contains both immanence and transcendence. (shrink)