We propose a study of the constitution of meaning in human-computer interaction based on Turing and Wittgenstein's definitions of thought, understanding, and decision. We show by the comparative analysis of the conceptual similarities and differences between the two authors that the common sense between humans and machines is co-constituted in and from action and that it is precisely in this co-constitution that lies the social value of their interaction. This involves problematizing human-machine interaction around the question of what it means to "follow a rule" to define and distinguish the interpretative modes and decision-making behaviors of each. We conclude that the mutualization of signs that takes place through the human-machine dialogue is at the foundation of the constitution of a computerized society.
相關內容
This study explores the capabilities of Large Language Models (LLMs), particularly OpenAI's ChatGPT, in addressing the challenges associated with software modeling, explicitly focusing on the bidirectional traceability problem between design models and code. The objective of this study is to demonstrate the proficiency of ChatGPT in understanding and integrating specific requirements into design models and code and its potential to offer solutions to the bidirectional traceability problem through a case study. The findings indicate that ChatGPT is capable of generating design models and code from natural language requirements, thereby bridging the gap between these requirements and software modeling. Despite its limitations in suggesting a specific method to resolve the problem using ChatGPT itself, it exhibited the capacity to provide corrections to be consistent between design models and code. As a result, the study concludes that achieving bidirectional traceability between design models and code is feasible using ChatGPT.
The emergence of Tiny Machine Learning (TinyML) has positively revolutionized the field of Artificial Intelligence by promoting the joint design of resource-constrained IoT hardware devices and their learning-based software architectures. TinyML carries an essential role within the fourth and fifth industrial revolutions in helping societies, economies, and individuals employ effective AI-infused computing technologies (e.g., smart cities, automotive, and medical robotics). Given its multidisciplinary nature, the field of TinyML has been approached from many different angles: this comprehensive survey wishes to provide an up-to-date overview focused on all the learning algorithms within TinyML-based solutions. The survey is based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodological flow, allowing for a systematic and complete literature survey. In particular, firstly we will examine the three different workflows for implementing a TinyML-based system, i.e., ML-oriented, HW-oriented, and co-design. Secondly, we propose a taxonomy that covers the learning panorama under the TinyML lens, examining in detail the different families of model optimization and design, as well as the state-of-the-art learning techniques. Thirdly, this survey will present the distinct features of hardware devices and software tools that represent the current state-of-the-art for TinyML intelligent edge applications. Finally, we discuss the challenges and future directions.
The energy efficiency of analog forms of computing makes it one of the most promising candidates to deploy resource-hungry machine learning tasks on resource-constrained system such as mobile or embedded devices. However, it is well known that for analog computations the safety net of discretization is missing, thus all analog computations are exposed to a variety of imperfections of corresponding implementations. Examples include non-linearities, saturation effect and various forms of noise. In this work, we observe that the ordering of input operands of an analog operation also has an impact on the output result, which essentially makes analog computations non-associative, even though the underlying operation might be mathematically associative. We conduct a simple test by creating a model of a real analog processor which captures such ordering effects. With this model we assess the importance of ordering by comparing the test accuracy of a neural network for keyword spotting, which is trained based either on an ordered model, on a non-ordered variant, and on real hardware. The results prove the existence of ordering effects as well as their high impact, as neglecting ordering results in substantial accuracy drops.
We provide the first convergence guarantee for full black-box variational inference (BBVI), also known as Monte Carlo variational inference. While preliminary investigations worked on simplified versions of BBVI (e.g., bounded domain, bounded support, only optimizing for the scale, and such), our setup does not need any such algorithmic modifications. Our results hold for log-smooth posterior densities with and without strong log-concavity and the location-scale variational family. Also, our analysis reveals that certain algorithm design choices commonly employed in practice, particularly, nonlinear parameterizations of the scale of the variational approximation, can result in suboptimal convergence rates. Fortunately, running BBVI with proximal stochastic gradient descent fixes these limitations, and thus achieves the strongest known convergence rate guarantees. We evaluate this theoretical insight by comparing proximal SGD against other standard implementations of BBVI on large-scale Bayesian inference problems.
6G promises a paradigm shift in which positioning and sensing are inherently integrated, enhancing not only the communication performance but also enabling location- and context-aware services. Historically, positioning and sensing have been viewed through the lens of cost and performance trade-offs, implying an escalated demand for resources, such as radio, physical, and computational resources, for improved performance. However, 6G goes beyond this traditional perspective to encompass a set of broader values, namely sustainability, inclusiveness, and trustworthiness. This paper aims to: (i) shed light on these important value indicators and their relationship with the conventional key performance indicators, and (ii) unveil the dual nature of 6G in relation to these key value indicators (i.e., ensuring operation according to the values and enabling services that affect the values).
We explore the space of matrix-generated (0, m, 2)-nets and (0, 2)-sequences in base 2, also known as digital dyadic nets and sequences. In computer graphics, they are arguably leading the competition for use in rendering. We provide a complete characterization of the design space and count the possible number of constructions with and without considering possible reorderings of the point set. Based on this analysis, we then show that every digital dyadic net can be reordered into a sequence, together with a corresponding algorithm. Finally, we present a novel family of self-similar digital dyadic sequences, to be named $\xi$-sequences, that spans a subspace with fewer degrees of freedom. Those $\xi$-sequences are extremely efficient to sample and compute, and we demonstrate their advantages over the classic Sobol (0, 2)-sequence.
This paper presents a comprehensive study focusing on the influence of DEM type and spatial resolution on the accuracy of flood inundation prediction. The research employs a state-of-the-art deep learning method using a 1D convolutional neural network (CNN). The CNN-based method employs training input data in the form of synthetic hydrographs, along with target data represented by water depth obtained utilizing a 2D hydrodynamic model, LISFLOOD-FP. The performance of the trained CNN models is then evaluated and compared with the observed flood event. This study examines the use of digital surface models (DSMs) and digital terrain models (DTMs) derived from a LIDAR-based 1m DTM, with resolutions ranging from 15 to 30 meters. The proposed methodology is implemented and evaluated in a well-established benchmark location in Carlisle, UK. The paper also discusses the applicability of the methodology to address the challenges encountered in a data-scarce flood-prone region, exemplified by Pakistan. The study found that DTM performs better than DSM at lower resolutions. Using a 30m DTM improved flood depth prediction accuracy by about 21% during the peak stage. Increasing the resolution to 15m increased RMSE and overlap index by at least 50% and 20% across all flood phases. The study demonstrates that while coarser resolution may impact the accuracy of the CNN model, it remains a viable option for rapid flood prediction compared to hydrodynamic modeling approaches.
The rapid development of deep learning has made a great progress in segmentation, one of the fundamental tasks of computer vision. However, the current segmentation algorithms mostly rely on the availability of pixel-level annotations, which are often expensive, tedious, and laborious. To alleviate this burden, the past years have witnessed an increasing attention in building label-efficient, deep-learning-based segmentation algorithms. This paper offers a comprehensive review on label-efficient segmentation methods. To this end, we first develop a taxonomy to organize these methods according to the supervision provided by different types of weak labels (including no supervision, coarse supervision, incomplete supervision and noisy supervision) and supplemented by the types of segmentation problems (including semantic segmentation, instance segmentation and panoptic segmentation). Next, we summarize the existing label-efficient segmentation methods from a unified perspective that discusses an important question: how to bridge the gap between weak supervision and dense prediction -- the current methods are mostly based on heuristic priors, such as cross-pixel similarity, cross-label constraint, cross-view consistency, cross-image relation, etc. Finally, we share our opinions about the future research directions for label-efficient deep segmentation.
This work considers the question of how convenient access to copious data impacts our ability to learn causal effects and relations. In what ways is learning causality in the era of big data different from -- or the same as -- the traditional one? To answer this question, this survey provides a comprehensive and structured review of both traditional and frontier methods in learning causality and relations along with the connections between causality and machine learning. This work points out on a case-by-case basis how big data facilitates, complicates, or motivates each approach.
We introduce a multi-task setup of identifying and classifying entities, relations, and coreference clusters in scientific articles. We create SciERC, a dataset that includes annotations for all three tasks and develop a unified framework called Scientific Information Extractor (SciIE) for with shared span representations. The multi-task setup reduces cascading errors between tasks and leverages cross-sentence relations through coreference links. Experiments show that our multi-task model outperforms previous models in scientific information extraction without using any domain-specific features. We further show that the framework supports construction of a scientific knowledge graph, which we use to analyze information in scientific literature.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191