Emma Brunskill

Building peer-to-peer systems with chord, a distributed lookup service

We argue that the core problem facing peer-to-peer Systems is locating documents in a decentralized network and propose Chord, a distributed lookup primitive. Chord provides an efficient method of locating documents while placing few constraints on the applications that use it. As proof that Chords functionality is useful in the development of peer-to-peer applications, we outline the implementation of a peer-to-peer file sharing system based on Chord.

Global and regional hearing impairment prevalence: an analysis of 42 studies in 29 countries

BACKGROUND Hearing impairment is a leading cause of disease burden, yet population-based studies that measure hearing impairment are rare. We estimate regional and global hearing impairment prevalence from sparse data and calculate corresponding uncertainty intervals. METHODS We accessed papers from a published literature review and obtained additional detailed data tabulations from investigators. We estimated the prevalence of hearing impairment by region, sex, age and hearing level using a Bayesian hierarchical model, a method that is effective for sparse data. As the primary objective of modelling was to produce regional and global prevalence estimates, including for those regions with scarce to no data, models were evaluated using cross-validation. RESULTS We used data from 42 studies, carried out between 1973 and 2010 in 29 countries. Hearing impairment was positively related to age, male sex and middle- and low-income regions. We estimated that the global prevalence of hearing impairment (defined as an average hearing level of 35 decibels or more in the better ear) in 2008 was 1.4% (95% uncertainty interval 1.0-2.2%) for children aged 5-14 years, 9.8% (7.7-13.2%) for females >15 years of age and 12.2% (9.7-16.2%) for males >15 years of age. The model exhibited good external validity in the cross-validation analysis, with 87% of survey estimates falling within our final models 95% uncertainty intervals. CONCLUSION Our results suggest that the prevalence of child and adult hearing impairment is substantially higher in middle- and low-income countries than in high-income countries, demonstrating the global need for attention to hearing impairment.

Designing mobile interfaces for novice and low-literacy users

While mobile phones have found broad application in bringing health, financial, and other services to the developing world, usability remains a major hurdle for novice and low-literacy populations. In this article, we take two steps to evaluate and improve the usability of mobile interfaces for such users. First, we offer an ethnographic study of the usability barriers facing 90 low-literacy subjects in India, Kenya, the Philippines, and South Africa. Then, via two studies involving over 70 subjects in India, we quantitatively compare the usability of different points in the mobile design space. In addition to text interfaces such as electronic forms, SMS, and USSD, we consider three text-free interfaces: a spoken dialog system, a graphical interface, and a live operator. Our results confirm that textual interfaces are unusable by first-time low-literacy users, and error prone for literate but novice users. In the context of healthcare, we find that a live operator is up to ten times more accurate than text-based interfaces, and can also be cost effective in countries such as India. In the context of mobile banking, we find that task completion is highest with a graphical interface, but those who understand the spoken dialog system can use it more quickly due to their comfort and familiarity with speech. We synthesize our findings into a set of design recommendations.

Topological mapping using spectral clustering and classification

In this work we present an online method for generating topological maps from raw sensor information. We first describe an algorithm to automatically decompose a map into submap segments using a graph partitioning technique known as spectral clustering. We then describe how to train a classifier to recognize graph submaps from laser signatures using the AdaBoost machine learning algorithm. We demonstrate that the we can perform topological mapping by incrementally segmenting the world as the robot moves through its environment, and we can close the loop when the learned classifier recognizes that the robot has returned to a previously visited location.

New Potentials for Data-Driven Intelligent Tutoring System Development and Optimization

Increasing widespread use of educational technologies is producing vast amounts of data. Such data can be used to help advance our understanding of student learning and enable more intelligent, interactive, engaging, and effective education. In this article, we discuss the status and prospects of this new and powerful opportunity for data-driven development and optimization of educational technologies, focusing on intelligent tutoring systems We provide examples of use of a variety of techniques to develop or optimize the select, evaluate, suggest, and update functions of intelligent tutors, including probabilistic grammar learning, rule induction, Markov decision process, classification, and integrations of symbolic search and statistical inference.

Efficient Planning under Uncertainty with Macro-actions

Deciding how to act in partially observable environments remains an active area of research. Identifying good sequences of decisions is particularly challenging when good control performance requires planning multiple steps into the future in domains with many states. Towards addressing this challenge, we present an online, forward-search algorithm called the Posterior Belief Distribution (PBD). PBD leverages a novel method for calculating the posterior distribution over beliefs that result after a sequence of actions is taken, given the set of observation sequences that could be received during this process. This method allows us to efficiently evaluate the expected reward of a sequence of primitive actions, which we refer to as macro-actions. We present a formal analysis of our approach, and examine its performance on two very large simulation experiments: scientific exploration and a target monitoring domain. We also demonstrate our algorithm being used to control a real robotic helicopter in a target monitoring experiment, which suggests that our approach has practical potential for planning in real-world, large partially observable domains where a multistep lookahead is required to achieve good performance.

SLAM using Incremental Probabilistic PCA and Dimensionality Reduction

The recent progress in robot mapping (or SLAM) algorithms has focused on estimating either point features (such as landmarks) or grid-based representations. Both of these representations generally scale with the size of the environment, not the complexity of the environment. Many thousand parameters may be required even when the structure of the environment can be represented using a few geometric primitives with many fewer parameters. We describe a novel SLAM model called IPSLAM. Our algorithm clusters sensor data into line segments using the Probabilistic PCA algorithm, which provides a data likelihood model that can be used within a SLAM algorithm for the simultaneous estimation of map and robot pose parameters. Unlike previous work in extracting line-based representations from point-based maps, IPSLAM builds non-point-based maps directly from the sensor data. We demonstrate our algorithm on mapping part of the MIT Stata Centre.

Faster teaching by POMDP planning

Both human and automated tutors must infer what a student knows and plan future actions to maximize learning. Though substantial research has been done on tracking and modeling student learning, there has been significantly less attention on planning teaching actions and how the assumed student model impacts the resulting plans. We frame the problem of optimally selecting teaching actions using a decision-theoretic approach and show how to formulate teaching as a partially-observable Markov decision process (POMDP) planning problem. We consider three models of student learning and present approximate methods for finding optimal teaching actions given the large state and action spaces that arise in teaching. An experimental evaluation of the resulting policies on a simple concept-learning task shows that framing teacher action planning as a POMDP can accelerate learning relative to baseline performance.

Faster Teaching via POMDP Planning

Human and automated tutors attempt to choose pedagogical activities that will maximize student learning, informed by their estimates of the students current knowledge. There has been substantial research on tracking and modeling student learning, but significantly less attention on how to plan teaching actions and how the assumed student model impacts the resulting plans. We frame the problem of optimally selecting teaching actions using a decision-theoretic approach and show how to formulate teaching as a partially observable Markov decision process planning problem. This framework makes it possible to explore how different assumptions about student learning and behavior should affect the selection of teaching actions. We consider how to apply this framework to concept learning problems, and we present approximate methods for finding optimal teaching actions, given the large state and action spaces that arise in teaching. Through simulations and behavioral experiments, we explore the consequences of choosing teacher actions under different assumed student models. In two concept-learning tasks, we show that this technique can accelerate learning relative to baseline performance.

Where to go: Interpreting natural directions using global inference

An important component of human-robot interaction is that people need to be able to instruct robots to move to other locations using naturally given directions. When giving directions, people often make mistakes such as labelling errors (e.g., left vs. right) and errors of omission (skipping important decision points in a sequence). Furthermore, people often use multiple levels of granularity in specifying directions, referring to locations using single object landmarks, multiple landmarks in a given location, or identifying large regions as a single location. The challenge is to identify the correct path to a destination from a sequence of noisy, possibly erroneous directions. In our work we cast this problem as probabilistic inference: given a set of directions, an agent should automatically find the path with the geometry and physical appearance to maximize the likelihood of those directions. We use a specific variant of a Markov Random Field (MRF) to represent our model, and gather multi-granularity representation information using existing large tagged datasets. On a dataset of route directions collected in a large third floor university building, we found that our algorithm correctly inferred the true final destination in 47 out of the 55 cases successfully followed by humans volunteers. These results suggest that our algorithm is performing well relative to human users. In the future this work will be included in a broader system for autonomously constructing environmental representations that support natural human-robot interaction for direction giving.