arXiv subjects

Deborah Estrin

Publications and source records attributed to Deborah Estrin.

14 recordsLinked to original sources

arXiv2026

Privacy Cards for Surfacing Mental Models and Exploring Privacy Concerns: A Case Study of Voice-First Ambient Interfaces with Older Adults

We investigate the ethical and privacy implications of voice-first ambient interfaces (VFAIs) for aging in place through an in-depth engagement with five older adults. Our participants were in the process of becoming experienced VFAI users, and had used a VFAI-based design probe for health data reporting. We create and iteratively refine an interview protocol using Privacy Cards. We customize Privacy Cards by drawing on participants' previous interviews and device usage logs. Using Privacy Cards, we conduct interviews to surface their mental models, and explore their privacy concerns. We find insufficient mental models for proper consent. For example, participants did not know who could access their data, and experienced difficulty distinguishing built-in functionality from third-party apps. Participants initially expressed little worry about VFAI-related ethical concerns, but interviews with Privacy Cards revealed nuanced issues, resulting in various implications for future research and design.

cs.HC↗

arXiv2024

Learning Disease Progression Models That Capture Health Disparities

Disease progression models are widely used to inform the diagnosis and treatment of many progressive diseases. However, a significant limitation of existing models is that they do not account for health disparities that can bias the observed data. To address this, we develop an interpretable Bayesian disease progression model that captures three key health disparities: certain patient populations may (1) start receiving care only when their disease is more severe, (2) experience faster disease progression even while receiving care, or (3) receive follow-up care less frequently conditional on disease severity. We show theoretically and empirically that failing to account for any of these disparities can result in biased estimates of severity (e.g., underestimating severity for disadvantaged groups). On a dataset of heart failure patients, we show that our model can identify groups that face each type of health disparity, and that accounting for these disparities while inferring disease severity meaningfully shifts which patients are considered high-risk.

cs.LG↗

arXiv2021

Towards Sparse Federated Analytics: Location Heatmaps under Distributed Differential Privacy with Secure Aggregation

We design a scalable algorithm to privately generate location heatmaps over decentralized data from millions of user devices. It aims to ensure differential privacy before data becomes visible to a service provider while maintaining high data accuracy and minimizing resource consumption on users' devices. To achieve this, we revisit distributed differential privacy based on recent results in secure multiparty computation, and we design a scalable and adaptive distributed differential privacy approach for location analytics. Evaluation on public location datasets shows that this approach successfully generates metropolitan-scale heatmaps from millions of user samples with a worst-case client communication overhead that is significantly smaller than existing state-of-the-art private protocols of similar accuracy.

cs.CR↗

arXiv2021

Policy-Based Federated Learning

In this paper we present PoliFL, a decentralized, edge-based framework that supports heterogeneous privacy policies for federated learning. We evaluate our system on three use cases that train models with sensitive user data collected by mobile phones - predictive text, image classification, and notification engagement prediction - on a Raspberry Pi edge device. We find that PoliFL is able to perform accurate model training and inference within reasonable resource and time budgets while also enforcing heterogeneous privacy policies.

cs.CR↗

arXiv2019

How To Backdoor Federated Learning

Federated learning enables thousands of participants to construct a deep learning model without sharing their private training data with each other. For example, multiple smartphones can jointly train a next-word predictor for keyboards without revealing what individual users type. We demonstrate that any participant in federated learning can introduce hidden backdoor functionality into the joint global model, e.g., to ensure that an image classifier assigns an attacker-chosen label to images with certain features, or that a word predictor completes certain sentences with an attacker-chosen word. We design and evaluate a new model-poisoning methodology based on model replacement. An attacker selected in a single round of federated learning can cause the global model to immediately reach 100% accuracy on the backdoor task. We evaluate the attack under different assumptions for the standard federated-learning tasks and show that it greatly outperforms data poisoning. Our generic constrain-and-scale technique also evades anomaly detection-based defenses by incorporating the evasion into the attacker's loss function during training.

cs.CR↗

arXiv2019

Creative Procedural-Knowledge Extraction From Web Design Tutorials

Complex design tasks often require performing diverse actions in a specific order. To (semi-)autonomously accomplish these tasks, applications need to understand and learn a wide range of design procedures, i.e., Creative Procedural-Knowledge (CPK). Prior knowledge base construction and mining have not typically addressed the creative fields, such as design and arts. In this paper, we formalize an ontology of CPK using five components: goal, workflow, action, command and usage; and extract components' values from online design tutorials. We scraped 19.6K tutorial-related webpages and built a web application for professional designers to identify and summarize CPK components. The annotated dataset consists of 819 unique commands, 47,491 actions, and 2,022 workflows and goals. Based on this dataset, we propose a general CPK extraction pipeline and demonstrate that existing text classification and sequence-to-sequence models are limited in identifying, predicting and summarizing complex operations described in heterogeneous styles. Through quantitative and qualitative error analysis, we discuss CPK extraction challenges that need to be addressed by future research.

cs.IR↗

arXiv2019

Augmenting Gastrointestinal Health: A Deep Learning Approach to Human Stool Recognition and Characterization in Macroscopic Images

Purpose - Functional bowel diseases, including irritable bowel syndrome, chronic constipation, and chronic diarrhea, are some of the most common diseases seen in clinical practice. Many patients describe a range of triggers for altered bowel consistency and symptoms. However, characterization of the relationship between symptom triggers using bowel diaries is hampered by poor compliance and lack of objective stool consistency measurements. We sought to develop a stool detection and tracking system using computer vision and deep convolutional neural networks (CNN) that could be used by patients, providers, and researchers in the assessment of chronic gastrointestinal (GI) disease.

cs.LG↗

arXiv2017

Yum-me: A Personalized Nutrient-based Meal Recommender System

Nutrient-based meal recommendations have the potential to help individuals prevent or manage conditions such as diabetes and obesity. However, learning people's food preferences and making recommendations that simultaneously appeal to their palate and satisfy nutritional expectations are challenging. Existing approaches either only learn high-level preferences or require a prolonged learning period. We propose Yum-me, a personalized nutrient-based meal recommender system designed to meet individuals' nutritional expectations, dietary restrictions, and fine-grained food preferences. Yum-me enables a simple and accurate food preference profiling procedure via a visual quiz-based user interface, and projects the learned profile into the domain of nutritionally appropriate food options to find ones that will appeal to the user. We present the design and implementation of Yum-me, and further describe and evaluate two innovative contributions. The first contriution is an open source state-of-the-art food image analysis model, named FoodDist. We demonstrate FoodDist's superior performance through careful benchmarking and discuss its applicability across a wide array of dietary applications. The second contribution is a novel online learning framework that learns food preference from item-wise and pairwise image comparisons. We evaluate the framework in a field study of 227 anonymous users and demonstrate that it outperforms other baselines by a significant margin. We further conducted an end-to-end validation of the feasibility and effectiveness of Yum-me through a 60-person user study, in which Yum-me improves the recommendation acceptance rate by 42.63%.

cs.HC↗

arXiv2016

Internet Scale Research Studies using SDL-RX

Medical research is one area where collecting data is usually hard and expensive. With the launch of ResearchKit, Apple and Sage Bionetworks made large-scale personal data collection increasingly popular via simple text-based survey apps running on mobile phones. But such surveys can be a barrier in terms of usability and richness of the data being collected. In this paper, we present SDL-R X , a powerful software library designed for ResearchKit that enables study-specific, personalized, and rich visual surveys, for both iOS and Android platforms.

cs.CY↗

arXiv2016

Your Activities of Daily Living (YADL): An Image-based Survey Technique for Patients with Arthritis

Healthcare professionals use Activities of Daily Living (ADL) to characterize a patient's functional status and to evaluate the effectiveness of treatment plans. ADLs are traditionally measured using standardized text-based questionnaires and the only form of personalization is in the form of question branching logic. Pervasive smartphone adoption makes it feasible to consider more frequent patient-reporting on ADLs. However, asking generic sets of questions repeatedly introduces user burden and fatigue that threatens to interfere with their utility. We introduce an approach called YADL (Your Activities of Daily Living) which uses images of ADLs and personalization to improve survey efficiency and the patient-experience. It offers several potential benefits: wider coverage of ADLs, improved engagement, and accurate capture of individual health situations. In this paper, we discuss our system design and the wide applicability of the design process for survey tools in healthcare and beyond. Interactions with with a small number of patients with Arthritis throughout the design process have been promising and we share detailed insights.

cs.CY↗

arXiv2015

Beyond Classification: Latent User Interests Profiling from Visual Contents Analysis

User preference profiling is an important task in modern online social networks (OSN). With the proliferation of image-centric social platforms, such as Pinterest, visual contents have become one of the most informative data streams for understanding user preferences. Traditional approaches usually treat visual content analysis as a general classification problem where one or more labels are assigned to each image. Although such an approach simplifies the process of image analysis, it misses the rich context and visual cues that play an important role in people's perception of images. In this paper, we explore the possibilities of learning a user's latent visual preferences directly from image contents. We propose a distance metric learning method based on Deep Convolutional Neural Networks (CNN) to directly extract similarity information from visual contents and use the derived distance metric to mine individual users' fine-grained visual preferences. Through our preliminary experiments using data from 5,790 Pinterest users, we show that even for the images within the same category, each user possesses distinct and individually-identifiable visual preferences that are consistent over their lifetime. Our results underscore the untapped potential of finer-grained visual preference profiling in understanding users' preferences.

cs.IR↗

arXiv2002

The STRESS Method for Boundary-point Performance Analysis of End-to-end Multicast Timer-Suppression Mechanisms

Evaluation of Internet protocols usually uses random scenarios or scenarios based on designers' intuition. Such approach may be useful for average-case analysis but does not cover boundary-point (worst or best-case) scenarios. To synthesize boundary-point scenarios a more systematic approach is needed.In this paper, we present a method for automatic synthesis of worst and best case scenarios for protocol boundary-point evaluation. Our method uses a fault-oriented test generation (FOTG) algorithm for searching the protocol and system state space to synthesize these scenarios. The algorithm is based on a global finite state machine (FSM) model. We extend the algorithm with timing semantics to handle end-to-end delays and address performance criteria. We introduce the notion of a virtual LAN to represent delays of the underlying multicast distribution tree. The algorithms used in our method utilize implicit backward search using branch and bound techniques and start from given target events. This aims to reduce the search complexity drastically. As a case study, we use our method to evaluate variants of the timer suppression mechanism, used in various multicast protocols, with respect to two performance criteria: overhead of response messages and response time. Simulation results for reliable multicast protocols show that our method provides a scalable way for synthesizing worst-case scenarios automatically. Results obtained using stress scenarios differ dramatically from those obtained through average-case analyses. We hope for our method to serve as a model for applying systematic scenario generation to other multicast protocols.

cs.NI↗

arXiv2000

Systematic Testing of Multicast Routing Protocols: Analysis of Forward and Backward Search Techniques

In this paper, we present a new methodology for developing systematic and automatic test generation algorithms for multipoint protocols. These algorithms attempt to synthesize network topologies and sequences of events that stress the protocol's correctness or performance. This problem can be viewed as a domain-specific search problem that suffers from the state space explosion problem. One goal of this work is to circumvent the state space explosion problem utilizing knowledge of network and fault modeling, and multipoint protocols. The two approaches investigated in this study are based on forward and backward search techniques. We use an extended finite state machine (FSM) model of the protocol. The first algorithm uses forward search to perform reduced reachability analysis. Using domain-specific information for multicast routing over LANs, the algorithm complexity is reduced from exponential to polynomial in the number of routers. This approach, however, does not fully automate topology synthesis. The second algorithm, the fault-oriented test generation, uses backward search for topology synthesis and uses backtracking to generate event sequences instead of searching forward from initial states. Using these algorithms, we have conducted studies for correctness of the multicast routing protocol PIM. We propose to extend these algorithms to study end-to-end multipoint protocols using a virtual LAN that represents delays of the underlying multicast distribution tree.

cs.NI↗

arXiv2000

Systematic Performance Evaluation of Multipoint Protocols

The advent of multipoint (multicast-based) applications and the growth and complexity of the Internet has complicated network protocol design and evaluation. In this paper, we present a method for automatic synthesis of worst and best case scenarios for multipoint protocol performance evaluation. Our method uses a fault-oriented test generation (FOTG) algorithm for searching the protocol and system state space to synthesize these scenarios. The algorithm is based on a global finite state machine (FSM) model. We extend the algorithm with timing semantics to handle end-to-end delays and address performance criteria. We introduce the notion of a virtual LAN to represent delays of the underlying multicast distribution tree. As a case study, we use our method to evaluate variants of the timer suppression mechanism, used in various multipoint protocols, with respect to two performance criteria: overhead of response messages and response time. Simulation results for reliable multicast protocols show that our method provides a scalable way for synthesizing worst-case scenarios automatically. We expect our method to serve as a model for applying systematic scenario generation to other multipoint protocols.

cs.NI↗