Alex `Sandy' Pentland

We use high-resolution mobile phone data with geolocation information and propose a novel technical framework to study how social influence propagates within a phone communication network and affects the offline decision to attend a performance event. Our fine-grained data are based on the universe of phone calls made in a European country between January and July 2016. We isolate social influence from observed and latent homophily by taking advantage of the rich spatial-temporal information and the social interactions available from the longitudinal behavioral data. We find that influence stemming from phone communication is significant and persists up to four degrees of separation in the communication network. Building on this finding, we introduce a new “influence” centrality measure that captures the empirical pattern of influence decay over successive connections. A validation test shows that the …

Online discourse faces challenges in facilitating substantive and productive political conversations. Recent technologies have explored the potential of generative AI to promote civil discourse, encourage the development of mutual understanding in a discussion, produce feedback that enables people to converge in their views, and provide usable citizen input on policy questions posed to the public by governments and civil society. In this paper, we present a framework to help policymakers, technologists, and the public assess potential opportunities and risks when incorporating generative AI into online platforms for discussion and deliberation in order to strengthen democratic practices and help democratic governments make more effective and responsive policy decisions.

Non-pharmaceutical measures such as preventive quarantines, remote working, school and workplace closures, lockdowns, etc. have shown effectiveness from an epidemic control perspective; however, they have also significant negative consequences on social life and relationships, work routines and community engagement. In particular, complex ideas, work and school collaborations, innovative discoveries and resilient norms formation and maintenance, which often require face-to-face interactions of two or more parties to be developed and synergically coordinated, are particularly affected. In this study, we propose an alternative hybrid solution that balances the slowdown of epidemic diffusion with the preservation of face-to-face interactions, that we test simulating a disease and a knowledge spreading simultaneously on a network of contacts. Our approach involves a two-step partitioning of the population …

AI systems are optimized to meet specific performance goals. Placing regulatory objectives among these goals allows AI systems to comply with Regulation by Design. The Regulation by Design approach has been explored with respect to privacy since the dawn of the Internet in the 1970s. Today, the rapid adoption of AI systems across industries allows Regulation by Design to be implemented much more broadly and it thus represents a valuable addition to the regulatory arsenal. In this chapter, we provide examples of how Regulation by Design might be implemented. Building on these use cases, we propose three important prerequisites for this regulatory approach–consensus on objectives, metrics for success, and auditing mechanisms. Finally, we explain the connection between Regulation by Design and Legal Dynamism, a related regulatory approach for algorithmic systems where success metrics are tied to external factors. We conclude by exploring some of the limitations of Regulation by Design.

Many studies have revealed the predictive power of the most frequent, regular and habitual mobility patterns. However, it remains unclear which components of the mobility patterns contain the most informative signals for predicting disparate economic development across urban areas. Here we use machine learning to predict economic outcomes by analyzing the heterogeneous mobility networks of 687 activities from more than 560,000 anonymized users in Boston, Chicago and Miami. We find that mobility patterns are highly predictive of the current and future economic development in major American cities but, surprisingly, the high predictive power is concentrated on infrequent, irregular and exploratory activities. These predictive activities account for only less than 2% of total visits but successfully explain more than 50% of variation in economic outcomes. Future research should shift more attention from regular …

The Internet has become a critical domain for modern society that requires ongoing efforts for its improvement and protection. Network traffic matrices are a powerful tool for understanding and analyzing networks and are broadly taught in online graph theory educational resources. Network traffic matrix concepts are rarely available in online computer network and cybersecurity educational resources. To fill this gap, an interactive game environment has been developed to teach the foundations of traffic matrices to the computer networking community. The game environment provides a convenient, broadly accessible, delivery mechanism that enables making material available rapidly to a wide audience. The core architecture of the game is a facility to add new network traffic matrix training modules via an easily editable JSON file. Using this facility an initial set of modules were rapidly created covering: basic traffic matrices, traffic patterns, security/defense/deterrence, a notional cyber attack, a distributed denial-of-service (DDoS) attack, and a variety of graph theory concepts. The game environment enables delivery in a wide range of contexts to enable rapid feedback and improvement. The game can be used as a core unit as part of a formal course or as a simple interactive introduction in a presentation.

How do skills shape career mobility and access to cities’ labor markets? Here we model career pathways as an occupation network constructed from the similarity of occupations’ skill requirements within each US city. Using a nationally representative survey and three resume datasets, skill similarity predicts transition rates between occupations and predictions improve with increasingly granular skill data. Thus, a measure for skill specialization based on a workers’ position in their city’s occupation network may predict future career dynamics. Job changes that decrease workers’ network embeddedness also increased wages, and workers tend to decrease their embeddedness over their careers. Further, city pairs with dissimilar job embeddedness have greater census migration and increased flows of enplaned passengers according to the US Bureau of Transportation Statistics. This study directly connects …

In a world of increasing closed-source commercial machine learning models, model evaluations from developers must be taken at face value. These benchmark results, whether over task accuracy, bias evaluations, or safety checks, are traditionally impossible to verify by a model end-user without the costly or impossible process of re-performing the benchmark on black-box model outputs. This work presents a method of verifiable model evaluation using model inference through zkSNARKs. The resulting zero-knowledge computational proofs of model outputs over datasets can be packaged into verifiable evaluation attestations showing that models with fixed private weights achieve stated performance or fairness metrics over public inputs. These verifiable attestations can be performed on any standard neural network model with varying compute requirements. For the first time, we demonstrate this across a sample of real-world models and highlight key challenges and design solutions. This presents a new transparency paradigm in the verifiable evaluation of private models.