Blocks

Blocks are the smallest pieces of svemir: links, images, notes and papers I’ve collected.

841 blocks · 77 channels · 841 nodes

'In Silver's vision, truly intelligent agents would need long time horizons. Chatbots have brief interactions: The user asks a question; the bot answers. But humans are forever conceiving long-term objectives, and planning what they need to do next week and next month in order to realize them. Future AIs, Silver believed, would behave in the same way. Taksed, for example, to help solve energy scarcity by inventing a superconductor, and AI might draw up a reading list, conduct experiments, invent novel materials, and so on, pursuing its goal over the space of a year or more. As Silver and Sutton wrote, the models of the future would "actively explore the world, adapt to changing environments, and discover strategies that might never occur to a human." "The kind of AI that we have today doesn't have a life. It doesn't have its own stream of experience in the way that an animal or human might have. And that needs to be changed, so that we can have systems that keep learning and learning. We'll have coding agents that are just there, continusly improvng worlds code, and predicting which tools you'll find more useful. They'll just be beavering away on all this stuff in the background. Or let's say you tell your agent that you want to learn a new skill - speaking Japanese, for example. It will go off and build an app for that. And then it will teach you in a way that's optimized for you. And you'd get better on some tests. And your improvement would reward the system, so that it learned how to be a better teacher even as you learned to be a better Japanese speaker.'
David Silver (DeepMind) on the future of AIFrom The Infinity Machine: Demis Hassabis, DeepMind and the Quest for Superintelligence by Sebastian Mallaby
It ’ s just distributed computing: Rethinking AI governanceIndia (TRAI · 2025What we now lump under the unitary label “artificial intelligence” is not a single technology, but a highly varied set of machine learning appli­ cations enabled and supported by a globally ubiquitous system of distributed computing. The paper introduces a 4 part conceptual framework for analyzing the structure of that system, which it labels the digital ecosystem. What we now call “AI” is then shown to be a general functionality of distributed computing. "AI” has been present in primitive forms from the origins of digital computing in the 1950s. Three short case studies show that large-scale machine learning applications have been present in the digital ecosystem ever since the rise of the Internet. and provoked the same public policy concerns that we now associate with “AI.” The governance problems of “AI” are really caused by the development of this digital ecosystem, not by LLMs or other recent applications of machine learning. The paper then examines five recent proposals to “govern AI”and maps them to the constituent elements of the digital ecosystem model. This mapping shows that real-world attempts to assert governance authority over AI capabilities requires systemic control of all four elements of the digital ecosystem: data, computing power, networks and software. “Governing AI,” in other words, means total control of distributed computing. A better alternative is to focus governance and regulation upon specific appli­ cations of machine learning. An application-specific approach to governance allows for a more decentralized, freer and more effective method of solving policy conflicts.
Advancements in Context Recognition for Edge Devices and Smart Eyewear: Sensors and ApplicationsFrancesca Palermo, Luca Casciano · 2025Edge devices have garnered significant attention for their ability to process data locally, providing low-latency, context-aware services without the need for extensive reliance on cloud computing. This capability is particularly crucial in context recognition, which enables dynamic adaptation to a user’s real-time environment. Applications range from health monitoring and augmented reality to smart assistance and social interaction analysis. Among edge devices, smart eyewear has emerged as a promising platform for context recognition due to its ability to unobtrusively capture rich, multi-modal sensor data. However, the deployment of context-aware systems on such devices presents unique challenges, including real- time processing, energy efficiency, sensor fusion, and noise management. This manuscript provides a comprehensive survey of context recognition in edge devices, with a specific emphasis on smart eyewear. It reviews the state-of-the-art sensors and applications for context inference. Furthermore, the paper discusses key challenges in achieving reliable, low-latency context recognition while addressing energy and computational constraints. By synthesizing advancements and identifying gaps, this work aims to guide the development of more robust and efficient solutions for context recognition in edge computing. INDEX TERMS Context recognition, edge computing, smarteyewear, wearable technology.
Identifying Hand-based Input Preference Based on Wearable EEGKaining Zhang∗ · 2024Understanding user input preference can improve the user experi- ence, however automatically determining preference can be diffi- cult. In this paper, we designed an EEG-based method for directly evaluating hand-based input preference for touch and mid-air ges- tures on a smartwatch. We conducted a two-phase experiment, recording EEG data from 18 participants as they performed ges- tures and captured their ratings (Phase 1) and preference choices (Phase 2) for each gesture. Our analysis uncovered distinct EEG patterns between preferred and non-preferred gestures, including significant differences in Power Spectral Density (PSD), Coherence (Coh), and Sample Entropy (SE) features. When participants en- gaged with their preferred input gestures, we identified decreased brain activity (PSD) in the central and occipital regions, reduced brain connectivity (Coh) in the delta and alpha bands, and increased brain complexity (SE) in multiple sizes. These insights offer the potential to develop rapid detection of user intent for interactive computing devices by analysing brain signals. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. AHs 2024, April 04–06, 2024, Melbourne, VIC, Australia © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM. ACM ISBN 979-8-4007-0980-7/24/04 https://doi.org/10.1145/3652920.3653028 CCS CONCEPTS • Human-centered computing →User studies; Gestural input; Laboratory experiments.
Recommendations user interface in web-based e-commerce systemsOleksandr Yeroshkin, Janusz Sobecki · 2024In the world of dynamic e-commerce, competition is fierce, and the ability to attract and retain customers is becoming crucial. In this reality, adaptive user interfaces (UI) play an important role, allowing e-commerce platforms to personalize the shopping experience to better understand and meet customer needs. When combined with advanced artificial intelligence (AI) technologies, adaptive UI can bring revolutionary changes to the way customers interact with e-commerce platforms. In this article, we will discuss how AI can be used to design and improve adaptive user interfaces in e-commerce systems, and what benefits can result from this combination. © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0](https://creativecommons.org/licenses/by-nc-nd/4.0)) Peer-review under responsibility of the scientific committee of the 28th International Conference on Knowledge Based and Intelligent information and Engineering Systems Oleksandr Yeroshkin et al. / Procedia Computer Science 246 (2024) 2874–2881 2875 tion of an experimental approach in a controlled environment. During the implementation and verification phases, we will be able to generate alternative solutions and evaluate them according to the UX and/or conversion requirements. The paper is structured as follows. In the second section, the state of the art is presented. Then, in the following section, an example of a commercial system whose user interface will be adapted is given. In the fourth section, the proposed solution of adapted UI based on AI is described. In the final section, we summarize the paper and give some information on the proposed future work. 2. State of the art In the beginning, UI recommendation services did not use artificial intelligence methods [12]. According to Baraglia [4] and Kopel [10], mainly traditional approaches were used at that time. However, currently we are dealing with solutions that intensively use various AI algorithms. Automating user interface optimization can be achieved using a variety of techniques based on both traditional ap- proaches and AI and machine learning (ML) algorithms. When it comes to optimizing user experience in e-commerce, there are three approaches: adaptable, semi-adaptive, and fully adaptive [1]. The first involves manual adjustment by a user or expert according to their usage criteria [18]. The second involves manual adjustment supported by system recommendations which is a kind of evol
Predictive Analytics for Website User BehaviorPawan Vankhede · 2024This study examines the application of predictive analytics to user behavior analysis on websites, with a particular emphasis on a dataset of customer behavior from e-commerce. To forecast behavioral patterns, the study will apply the random forest method. Additionally, exploratory data analysis (EDA) will be carried out to extract insights from the dataset. The dataset employed in this study includes a variety of factors, such as purchase history, browsing habits, and consumer demographics, among other pertinent data. The research attempts to derive useful insights that can improve the comprehension and prediction of user behavior on the website by utilizing predictive analytics approaches. In this study, the random forest algorithm which is well-known for its capacity to manage intricate datasets and generate precise predictions is employed. The method builds several decision trees and combines their predictions to produce more reliable and accurate outcomes by utilizing ensemble learning. Predicting different aspects of user behavior, including browsing patterns, chance of making a purchase, and customer segmentation, is the aim. When paired with the insights from EDA, the outcomes of the behavioral analysis prediction made with the random forest algorithm can provide e-commerce companies with useful data. Gaining insight into user behavior can help with customer engagement, website design optimization, personalization, and overall user experience enhancement.
Educating Artificial Intelligence following the Child Learning Development Trajectories2024Artificial Intelligence is spreading in most daily activities. However, its develop- ment and deployment raise issues related to biases, such as gender and disability, mainly stemming from biased or incomplete datasets and lack of transparency and accountability in its algorithms. To overcome these issues, it is necessary to revert to a human-centered mindset, trying to educate algorithms rather than only train them. Adopting a human-centered approach in AI has been a first step, but it is necessary a step ahead. Indeed, recent theoretical perspectives suggest that edu- cating AI algorithms also need a profound understanding of the context of use in which it operates, adopting an approach like those in which a child is educated from birth following its developmental trajectory. By incorporating well-established educational models into the training of AI algorithms, intelligent systems based on those AI algorithms can better align with human learning trajectories, reducing bi- ases and making them more contextually aware. This paper goes in this direction, presenting an educational human-centered approach as a design methodology for artificial intelligence algorithms used within the European FRACTAL project. This proposal would pave the way toward developing more educated artificial intelligence algorithms since they are adapted to the real context of use.
Harnessing AI potential in E-Commerce: improving user engagement and sales through deep learning-based product recommendationsZhang · 2024Artificial intelligence (AI) has become a game-changing influence in e-commerce, reshaping the way businesses inter­ act with customers and increasing operational efficiency. In today’s digital era, AI technologies are being progressively embedded into e-commerce platforms to enhance the user experience, improve processes, and drive business growth. One of the primary areas where AI demonstrates its potential is in personalized product recommendations. This study explores the development of AI-driven recommendations regarding product design, sales, and customer experience in e-commerce environments. The research employs a quantitative approach to explore how AI technologies can enhance various facets of online retail. Using purposive sampling, 439 consumers and 356 sellers from diverse regions of China participated in an online survey designed to gather insights into their interactions with AI-driven recommendations. Data analysis was rigorously conducted using Lisrel 10.20 and Smart PLS 3 software, focusing on validating hypotheses related to AI’s influence on consumer behavior and business outcomes. Key findings include AI’s significant role in boosting productiv­ ity, enhancing sales figures, and enriching consumer experiences through personalized recommendations. This research provides a deeper insight into AI’s potential in e-commerce and provides recommendations for businesses focused on using AI technology to increase user engagement and increase sales in digital marketing.
Collaborative Filtering and Recommendation Algorithm for Artificial Intelligence Live Streaming E-Commerce Platforms Based on Big DataChen · 2024With the development of the Internet and mobile Internet, live streaming e-commerce has become an emerging e-commerce force. However, traditional recommendation algorithms have shortcomings in terms of accuracy and personalization of recommendation results, and more intelligent and personalized recommendation algorithms need to be applied. This article aimed to achieve personalized product recommendations and enhance the shopping experience of users by analyzing their historical behavioral data, real-time interests and needs, combined with big data and artificial intelligence technology. The collaborative filtering recommendation algorithm based on live streaming had an average recommendation accuracy of over 80% for user groups 1, 2, and 3. The research results of this article had important practical significance for promoting the healthy development of live streaming e-commerce platforms, improving user experience, and enhancing platform competitiveness. © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0](https://creativecommons.org/licenses/by-nc-nd/4.0)) Peer-review under responsibility of the scientific committee of the 11th International Conference on Applications and Techniques in Cyber Intelligence Yu’e Chen et al. / Procedia Computer Science 247 (2024) 826–833 827 and loyalty to the platform, but also promotes user purchasing behavior, increases purchase conversion rates, and enhances order value. This article analyzes the current application status of traditional recommendation algorithms and big data artificial intelligence in the e-commerce field, points out the shortcomings and areas for improvement of existing methods, and proposes methods such as user collaborative filtering and live streaming collaborative filtering to improve recommendation accuracy. On this basis, an in-depth analysis and exploration of collaborative filtering and recommendation algorithms for intelligent live streaming e-commerce platforms based on big data is conducted. 2. Related Works A large number of customers use traditional e-commerce portal websites, which lack product quality assurance. Rashidin Md Salamun used the status quo bias theory to study customer retention behavior on e-commerce platforms. The research findings can help managers and policy makers develop new policies to better serve customers [1]. Nichifor Eliza aimed to study the impact of using artificial intelligence through chatbots on the con
designed & built by Tanja Radovanovic