Federated learning is one of the main research lines in the last years about distributed learning, where participating nodes share their models but maintain the privacy of the data used to learn such models. Consensus...
详细信息
ISBN:
(纸本)9783031777370;9783031777387
Federated learning is one of the main research lines in the last years about distributed learning, where participating nodes share their models but maintain the privacy of the data used to learn such models. Consensus is a way of calculating a mean value between a set of agents using only information from the local neighbors. this paper presents a new approach based on Asynchronous Consensus, called Multi-layered Asynchronous Consensus-based Federated learning (MACoFL). It randomly chooses a neighbor and a layer from the neural model and interchanges it with him. this new algorithm is presented and tested using the MNIST dataset.
the proceedings contain 12 papers. the special focus in this conference is on Fundamentals of Software engineering. the topics include: automated Test Generation: Taxonomy and Tool Applications;finding Universally Qua...
ISBN:
(纸本)9783031870538
the proceedings contain 12 papers. the special focus in this conference is on Fundamentals of Software engineering. the topics include: automated Test Generation: Taxonomy and Tool Applications;finding Universally Quantified Heap Invariants by Horn Clause Transformations;a Framework for Model-Based Specification and Verification in Feature-Oriented Software Product Lines;extracting Formal Models for User’s Behaviors in Social Networks Using Automata and Machine learning;data-Driven Shielding of Online Reinforcement learning: A Stormwater Pond Case Study;on Explicit Solutions to Fixed-Point Equations in Propositional Dynamic Logic;on Time-Sensitive Control Closure for Secure Information Flow;automatic Generation of Loop Invariants in Dafny with Large Language Models;streamlining Parameter Tuning in Full-Body Racing Simulators with an automated Pipeline;formally Verified Verifiable Group Generators.
作者:
Mitic, PeterUCL
Dept Comp Sci Gower St London WC1E 6BT England
We present a quantitative definition of reputation risk, formulated in terms of a reputation time series comprising daily sentiment measurements. Self Supported learning is used to quantify reputation risk by progress...
详细信息
ISBN:
(纸本)9783031777301;9783031777318
We present a quantitative definition of reputation risk, formulated in terms of a reputation time series comprising daily sentiment measurements. Self Supported learning is used to quantify reputation risk by progressively refining an initial proposal for a Minimum Acceptable Sentiment, calculated from descriptive statistics of the reputation data. the derived values are validated using a "sense test" based on a Loess quantile. the results show that the Minimum Acceptable Sentiment value is given approximately by a two standard deviation lower tail of the observed data.
this paper presents a novel deep-learning pipeline to segment large railway datasets with minimal manual annotation, notoriously time consuming. the pipeline adapts DINOv2 [11] for labeling point clouds, with tailored...
详细信息
ISBN:
(纸本)9783031777301;9783031777318
this paper presents a novel deep-learning pipeline to segment large railway datasets with minimal manual annotation, notoriously time consuming. the pipeline adapts DINOv2 [11] for labeling point clouds, with tailored self-distillation pre-training and fine-tuning. the adopted transformer architecture successfully generalizes to multiple railway datasets, with a lightweight pipeline that outperforms manual labeling speed by a factor of 6, despite requiring a final segmentation check and correction. this groundbreaking achievement bridges the gap between the need for annotated point clouds in railway industry and the lack of publicly available annotated datasets.
Systems based on artificial intelligence have become prominent in nearly all domains. However, knowledge of the inner workings of these intelligent systems is not as widespread, partly because the associated issues ha...
详细信息
ISBN:
(纸本)9783031777370;9783031777387
Systems based on artificial intelligence have become prominent in nearly all domains. However, knowledge of the inner workings of these intelligent systems is not as widespread, partly because the associated issues have been discussed only to a limited extent in computer science education. In order to gain an overview of AI in curricula and to see what competencies teachers need to teach this content, the AIrelated content of the computer science curricula of the German federal states was analysed and compared with existing approaches. Proposals for further training courses are derived from this to enable teachers to teach AI competently.
An accurate and reliable image-based fruit detection system is essential for advancing agricultural tasks such as yield mapping and robotic harvesting. this paper benchmarks five state-of-the-art object detection fram...
详细信息
Online hotel booking became increasingly popular as time passed, and with its popularity, the datathat can be collected based on customer actions has increased. this data can serve to build intelligent systems that c...
详细信息
ISBN:
(纸本)9783031777370;9783031777387
Online hotel booking became increasingly popular as time passed, and with its popularity, the datathat can be collected based on customer actions has increased. this data can serve to build intelligent systems that can provide knowledge for both customers and hotel owners. In this paper, we focus on hotel owners who can benefit from the collected data by adjusting the prices to optimise the profit of their accommodations. To accomplish this, we built a system that collected the data from *** and gathered a helpful dataset for price prediction. We used five regression algorithms and an optimization technique to obtain the best results, leading us to a 9% error for price prediction. this result allows accommodation owners to predict the room price to keep the rooms fully occupied.
this study introduces a novel framework for the automatic two-dimensional tracking of padel games using monocular recordings. By integrating advanced Computer Vision and Deep learning techniques, our algorithm detects...
详细信息
ISBN:
(纸本)9783031777301;9783031777318
this study introduces a novel framework for the automatic two-dimensional tracking of padel games using monocular recordings. By integrating advanced Computer Vision and Deep learning techniques, our algorithm detects and tracks players, the court, and the ball. through homography, we accurately project detected player positions onto a two-dimensional court, enabling comprehensive tracking throughout the game. We tested the proposed algorithm using amateur video recordings of padel games found in literature. this approach remains user-friendly, cost-effective, and adaptable to various camera angles and lighting conditions. this makes it accessible to both amateur and professional players and coaches, providing a valuable tool for performance analysis. Additionally, the proposed framework holds potential for adaptation to other sports with minimal modifications, further broadening its applicability.
When delivered to the market, machine learning models face new data which are possibly subject to novel characteristics - a phenomenon known as concept drift. As this might lead to performance degradation, it is neces...
详细信息
ISBN:
(纸本)9783031777301;9783031777318
When delivered to the market, machine learning models face new data which are possibly subject to novel characteristics - a phenomenon known as concept drift. As this might lead to performance degradation, it is necessary to detect such drift and, if required, adapt the model accordingly. While a variety of drift detection and adaptation methods exists for standard vectorial data, a suitable treatment of text data is less researched. In this work we present a novel approach which detects and explains drift in text data based on their representation via transformer embeddings. In a nutshell, the method generates suitable statistical features from the original distribution and the possibly shifted variation. Based on these representations, drift scores can be assigned to individual data points, allowing a visualization and human-readable characterization of the type of drift. We demonstrate the approach's effectiveness in reliably detecting drift in several experiments.
Federated learning (FL) is a prominent method in machine learning, that ensures privacy by enabling distributed devices to collaboratively learn a shared model without exchanging local data. this paper provides a comp...
详细信息
ISBN:
(纸本)9783031777370;9783031777387
Federated learning (FL) is a prominent method in machine learning, that ensures privacy by enabling distributed devices to collaboratively learn a shared model without exchanging local data. this paper provides a comparative analysis of various FL algorithms implemented on the Smart Python Agent Development Environment (SPADE) framework. We focus on evaluating the performance, scalability, and resilience of these algorithms across different network setups and data distribution scenarios. Our results highlight the differential impacts of decentralized versus centralized approaches, particularly under non-IID data conditions, common in real-world applications. By leveraging SPADE agents and consensus algorithms, this study not only tests algorithmic efficiency and system robustness but also explores advanced strategies like asynchronous updates and coalition-based learning, which show promise in enhancing model accuracy and reducing communication overhead.
暂无评论