Yaniv Altshuler

Cooperative Cleaners: A Study in Ant Robotics

In the world of living creatures, simple-minded animals often cooperate to achieve common goals with amazing performance. One can consider this idea in the context of robotics, and suggest models for programming goal-oriented behavior into the members of a group of simple robots lacking global supervision. This can be done by controlling the local interactions between the robot agents, to have them jointly carry out a given mission. As a test case we analyze the problem of many simple robots cooperating to clean the dirty floor of a non-convex region in Z 2 , using the dirt on the floor as the main means of inter-robot communication.

Workforce optimization: identification and assignment of professional workers using constraint programming

Matching highly skilled people to available positions is a high-stakes task that requires careful consideration by experienced resource managers. A wrong decision may result in significant loss of value due to understaffing, underqualification or overqualification of assigned personnel, and high turnover of poorly matched workers. While the importance of quality matching is clear, dealing with pools of hundreds of jobs and resources in a dynamic market generates a significant amount of pressure to make decisions rapidly. We present a novel solution designed to bridge the gap between the need for high-quality matches and the need for timeliness. By applying constraint programming, a subfield of artificial intelligence, we are able to deal successfully with the complex constraints encountered in the field and reach near-optimal assignments that take into account all resources and positions in the pool. The considerations include constraints on job role, skill level, geographical location, language, potential retraining, and many more. Constraints are applied at both the individual and team levels. This paper introduces the technology and then describes its use by IBM Global Services, where large numbers of service and consulting employees are considered when forming teams assigned to customer projects.

Efficient cooperative search of smart targets using uav swarms1

This work examines the Cooperative Hunters problem, where a swarm of unmanned air vehicles (UAVs) is used for searching one or more “evading targets,” which are moving in a predefined area while trying to avoid a detection by the swarm. By arranging themselves into efficient geometric flight configurations, the UAVs optimize their integrated sensing capabilities, enabling the search of a maximal territory.

Augmented Betweenness Centrality for Environmentally Aware Traffic Monitoring in Transportation Networks

Network planning and traffic flow optimization require the acquisition and analysis of large quantities of data such as the network topology, its traffic flow data, vehicle fleet composition, emission measurements and so on. Data acquisition is an expensive process that involves household surveys and automatic as well as semiautomatic measurements performed all over the network. For example, in order to accurately estimate the effect of a certain network change on the total emissions produced by vehicles in the network, assessment of the vehicle fleet composition for each origin–destination pair is required. As a result, problems that optimize nonlocal merit functions become highly difficult to solve. One such problem is finding the optimal deployment of traffic monitoring units. In this article we suggest a new traffic assignment model that is based on the concept of shortest path betweenness centrality measure, borrowed from the domain of complex network analysis. We show how betweenness can be augmented in order to solve the traffic assignment problem given an arbitrary travel cost definition. The proposed traffic assignment model is evaluated using a high-resolution Israeli transportation data set derived from the analysis of cellular phones data. The group variant of the augmented betweenness centrality is then used to optimize the locations of traffic monitoring units, hence reducing the cost and increasing the effectiveness of traffic monitoring.

Stealing Reality: When Criminals Become Data Scientists (or Vice Versa)

Stealing-reality attacks attempt to steal social network and behavioral information through data collection and inference techniques, making them more dangerous than other types of identity theft.

Multi-agent Cooperative Cleaning of Expanding Domains

Several recent works considered multi-a(ge)nt robotics in static environments. In this work we examine ways of operating in dynamic environments, where changes take place independently of the agents’ activity. The work focuses on a dynamic variant of the cooperative cleaners problem, a problem that requires several simple agents to clean a connected region of “dirty” pixels in Z 2 . A number of simple agents move in this dirty region, each having the ability to “clean” the place it is located in. Their goal is to jointly clean the given dirty region. The dynamic variant of the problem involves a deterministic expansion of dirt in the environment, simulating spreading of contamination or fire. Theoretical lower bounds for the problem are presented, as well as various impossibility results. A cleaning protocol for the problem is presented, and a wealth of experimental results testing its performance in comparison to the lower bounds. Several analytic upper bounds for the proposed protocol are also presented, accompanied with appropriate experimental results.

Incremental Learning with Accuracy Prediction of Social and Individual Properties from Mobile-Phone Data

As truly ubiquitous wearable computers, mobile phones are quickly becoming the primary source for social, behavioral, and environmental sensing and data collection. Todays smart phones are equipped with increasingly more sensors and accessible data types that enable the collection of literally dozens of signals regarding the phone, its user, and their environment. A great deal of research effort in academia and industry is put into mining this data for higher level sense-making, such as understanding user context, inferring social networks, learning individual features, and so on. In many cases this analysis work is the result of exploratory forays and trial-and-error. Adding to the challenge, the devices themselves are limited platforms, hence data collection campaign must be carefully designed in order to collect the signals in the appropriate frequency, avoiding the exhausting the the devices limited battery and processing power. Currently however, there is no structured methodology for the design of mobile data collection and analysis initiatives. In this work we investigate the properties of learning and inference of real world data collected via mobile phones over time. In particular, we analyze how the ability to predict individual parameters and social links is incrementally enhanced with the accumulation of additional data. To do so we use the Friends and Family dataset, containing rich data signals gathered from the smart phones of 140 adult members of an MIT based young-family residential community for over a year, and is one of the most comprehensive mobile phone datasets gathered in academia to date. We develop several models for predicting social and individual properties from sensed mobile phone data over time, including detection of life-partners, ethnicity, and whether a person is a student or not. Finally, we propose a method for predicting the maximal learning accuracy possible for the learning task at hand, based on an initial set of measurements. This has various practical implications, such as better design of mobile data collection campaigns, or evaluating of planned analysis strategies.

How many makes a crowd? on the evolution of learning as a factor of community coverage

As truly ubiquitous wearable computers, mobile phones are quickly becoming the primary source for social, behavioral and environmental sensing and data collection. Todays smartphones are equipped with increasingly more sensors and accessible data types that enable the collection of literally dozens of signals related to the phone, its user, and its environment. A great deal of research effort in academia and industry is put into mining this raw data for higher level sense-making, such as understanding user context, inferring social networks, learning individual features, and so on. In many cases, this analysis work is the result of exploratory forays and trial-and-error. In this work we investigate the properties of learning and inferences of real world data collected via mobile phones for different sizes of analyzed networks. In particular, we examine how the ability to predict individual features and social links is incrementally enhanced with the accumulation of additional data. To accomplish this, we use the Friends and Family dataset, which contains rich data signals gathered from the smartphones of 130 adult members of a young-family residential community over the course of a year and consequently has become one of the most comprehensive mobile phone datasets gathered in academia to date. Our results show that features such as ethnicity, age and marital status can be detected by analyzing social and behavioral signals. We then investigate how the prediction accuracy is increased when the users sample set grows. Finally, we propose a method for advanced prediction of the maximal learning accuracy possible for the learning task at hand, based on an initial set of measurements. These predictions have practical implications, such as influencing the design of mobile data collection campaigns or evaluating analysis strategies.

Swarm Intelligence — Searchers, Cleaners and Hunters

This chapter examines the concept of swarm intelligence through three examples of complex problems which are solved by a decentralized swarm of simple agents. The protocols employed by these agents are presented, as well as various analytic results for their performance and for the problems in general. The problems examined are the problem of finding patterns within physical graphs (e.g. k-cliques), the dynamic cooperative cleaners problem, and a problem concerning a swarm of UAVs (unmanned air vehicles), hunting an evading target (or targets). In addition, the work contains a discussion regarding open questions and ongoing and future research in this field.

Prospect Theory for Online Financial Trading

Prospect theory is widely viewed as the best available descriptive model of how people evaluate risk in experimental settings. According to prospect theory, people are typically risk-averse with respect to gains and risk-seeking with respect to losses, known as the “reflection effect”. People are much more sensitive to losses than to gains of the same magnitude, a phenomenon called “loss aversion”. Despite of the fact that prospect theory has been well developed in behavioral economics at the theoretical level, there exist very few large-scale empirical studies and most of the previous studies have been undertaken with micro-panel data. Here we analyze over 28.5 million trades made by 81.3 thousand traders of an online financial trading community over 28 months, aiming to explore the large-scale empirical aspect of prospect theory. By analyzing and comparing the behavior of winning and losing trades and traders, we find clear evidence of the reflection effect and the loss aversion phenomenon, which are essential in prospect theory. This work hence demonstrates an unprecedented large-scale empirical evidence of prospect theory, which has immediate implication in financial trading, e.g., developing new trading strategies by minimizing the impact of the reflection effect and the loss aversion phenomenon. Moreover, we introduce three novel behavioral metrics to differentiate winning and losing traders based on their historical trading behavior. This offers us potential opportunities to augment online social trading where traders are allowed to watch and follow the trading activities of others, by predicting potential winners based on their historical trading behavior.