ETH Zurich

About: ETH Zurich is a education organization based out in Zurich, Switzerland. It is known for research contribution in the topics: Population & Computer science. The organization has 48393 authors who have published 122408 publications receiving 5111383 citations. The organization is also known as: Swiss Federal Institute of Technology in Zurich & Eidgenössische Technische Hochschule Zürich.

A revised approach to water footprinting to make transparent the impacts of consumption and production on global freshwater scarcity

Abstract: Through the interconnectedness of global business, the local consumption of products and services is intervening in the hydrological cycle throughout the world to an unprecedented extent. In order to address the unsustainable use of global freshwater resources, indicators are needed which make the impacts of production systems and consumption patterns transparent. In this paper, a revised water footprint calculation method, incorporating water stress characterisation factors, is presented and demonstrated for two case study products, Dolmio ® pasta sauce and Peanut M&M's ® using primary production data. The method offers a simple, yet meaningful way of making quantitative comparisons between products, production systems and services in terms of their potential to contribute to water scarcity. As such, capacity is created for change through public policy as well as corporate and individual action. This revised method represents an alternative to existing volumetric water footprint calculation methods which combine green and blue water consumption from water scarce and water abundant regions such that they give no clear indication about where the actual potential for harm exists.

Temporal Segment Networks for Action Recognition in Videos

Abstract: We present a general and flexible video-level framework for learning action models in videos. This method, called temporal segment network (TSN), aims to model long-range temporal structure with a new segment-based sampling and aggregation scheme. This unique design enables the TSN framework to efficiently learn action models by using the whole video. The learned models could be easily deployed for action recognition in both trimmed and untrimmed videos with simple average pooling and multi-scale temporal window integration, respectively. We also study a series of good practices for the implementation of the TSN framework given limited training samples. Our approach obtains the state-the-of-art performance on five challenging action recognition benchmarks: HMDB51 (71.0 percent), UCF101 (94.9 percent), THUMOS14 (80.1 percent), ActivityNet v1.2 (89.6 percent), and Kinetics400 (75.7 percent). In addition, using the proposed RGB difference as a simple motion representation, our method can still achieve competitive accuracy on UCF101 (91.0 percent) while running at 340 FPS. Furthermore, based on the proposed TSN framework, we won the video classification track at the ActivityNet challenge 2016 among 24 teams.

Smartphone use and smartphone addiction among young people in Switzerland

Abstract: Background and AimsSmartphone addiction, its association with smartphone use, and its predictors have not yet been studied in a European sample. This study investigated indicators of smartphone use, smartphone addiction, and their associations with demographic and health behaviour-related variables in young people.MethodsA convenience sample of 1,519 students from 127 Swiss vocational school classes participated in a survey assessing demographic and health-related characteristics as well as indicators of smartphone use and addiction. Smartphone addiction was assessed using a short version of the Smartphone Addiction Scale for Adolescents (SAS-SV). Logistic regression analyses were conducted to investigate demographic and health-related predictors of smartphone addiction.ResultsSmartphone addiction occurred in 256 (16.9%) of the 1,519 students. Longer duration of smartphone use on a typical day, a shorter time period until first smartphone use in the morning, and reporting that social networking was the mo...

ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Models

Abstract: Machine learning (ML) has become a core component of many real-world applications and training data is a key factor that drives current progress. This huge success has led Internet companies to deploy machine learning as a service (MLaaS). Recently, the first membership inference attack has shown that extraction of information on the training set is possible in such MLaaS settings, which has severe security and privacy implications. However, the early demonstrations of the feasibility of such attacks have many assumptions on the adversary, such as using multiple so-called shadow models, knowledge of the target model structure, and having a dataset from the same distribution as the target model's training data. We relax all these key assumptions, thereby showing that such attacks are very broadly applicable at low cost and thereby pose a more severe risk than previously thought. We present the most comprehensive study so far on this emerging and developing threat using eight diverse datasets which show the viability of the proposed attacks across domains. In addition, we propose the first effective defense mechanisms against such broader class of membership inference attacks that maintain a high level of utility of the ML model.