Sanket Rajeev Sabharwal, PhD

Hi, this is Sanket.

I am a computer scientist and software engineer based in Italy. Help founders and C-suite executives solve business problems with the flexibility, scalability, and sheer beauty of software engineering and data, combined. Research in the domain of algorithmic-led HCI applications.

At this stage in my life I cherish reading extensively, particularly philosophy, annual reports and shareholder letters. My moments of 'being' involve indulging in physically demanding choices, such as Muay Thai and open-sea swimming, and intellectually demanding desires such as discussions on non-duality.

Experience

Education

Scientific Publications

Link: Click here

Abstract:
Music ensemble performance provides an ecologically valid context for investigating leadership dynamics in small group interactions. Musical texture, specifically the relative salience of simultaneously sounding ensemble parts, is a feature that can potentially alter leadership dynamics by introducing hierarchical relationships between individual parts. The present study extended previous work on quantifying interpersonal coupling in musical ensembles by examining the relationship between musical texture and leader-follower relations, operationalised as directionality of influence between co-performers’ body motion in concert video recordings. It was hypothesised that the directionality of influence, indexed by Granger Causality, would be greater for ‘homophonic’ textures with a clear distinction between melody and accompaniment parts than for ‘polyphonic’ textures with less distinction between melody and accompaniment. This hypothesis was tested by using pose estimation algorithms to track instrumentalists’ body movements in a string quartet and a clarinet quintet, and then applying Granger Causality analysis to their head motion to estimate directional influence between instrumentalist pairs for sections of the pieces that varied in texture. It was found that Granger Causality values were generally higher (indicating greater directionality of influence) for homophonic than polyphonic textures. Furthermore, considering melody and accompaniment instrument roles revealed more evidence for the melody instrument influencing accompanying instruments than vice versa, plus a high degree of directionality among accompanying instruments, in homophonic textures. These observed patterns of directional information flow in co-performer body motion are consistent with changing leader-follower relations depending on hierarchical relations between ensemble parts in terms of the relative salience of melodic material in the musical texture. The finding that automatic pose estimation can detect modulations of leadership dynamics in standard video recordings under naturalistic performance conditions has implications for investigating interpersonal coordination in large-scale music video datasets representing different cultural traditions, and for exploring nonverbal communication in group activities more generally.

Link: Click here

Abstract:
This doctoral thesis introduces huSync, a computational framework developed to assess non-verbal communication dynamics in small groups quantitatively. It aspires to contribute to Interaction Design and Human-Computer Interaction (HCI), focused on bridging computer science techniques, embodied design and somatics. huSync employs pose estimation algorithms to interpret movement trajectories from video sequences, offering a non-intrusive way to study entrainment in small-group settings based on established conceptual frameworks. Joint actions in musical ensembles serve as the primary case study to explore how non-verbal body cues particularly influence interpersonal coordination and the directionality of information flow. These musical joint actions are exemplary instances of 'self-managed groups', illustrating the complex relationships between musical structure, entrainment dynamics, and mutual influence among ensemble members. These musical interactions are central to the research, serving as a universal language to understand nuanced human behaviours. Methods and results derived from three distinct studies on dyadic group dynamics are also presented. Recent studies are making improvements in computationally modelling human behavior, sharing interesting techniques and approaches, and continues to remain an open area for research. huSync provides a computational methodology and approach to model these behavioural mechanisms in small-group setups, and in this thesis, we present musical ensembles as a use case. By studying these subtle interactions in group settings, we begin to get a clearer perspective on how groups connect and interact. The potential of huSync lies in its capability to make these intangible elements tangible, offering a fresh perspective on the subtleties of small-group interactions. huSync is versatile, managing diverse data sources and identifying essential movement attributes characteristic of group interactions, combining multi-modal signals, feature extraction, entrainment measurement, and analysis validation. It extends its applications to healthcare projects, emphasizing music's universal role in non-verbal communication and interdisciplinary studies by integrating technology with economics, psychology, and the arts. It aims to bridge diverse disciplines, suggesting new paths for research in human movement sciences, especially regarding the use of markerless technologies in behaviorally driven computational research. The findings from huSync are reliable and provide an alternative means for analysing human body movements, aiding in deepening the understanding of small-group dynamics and the elements contributing to successful collaborations. huSync seeks to provide dependable insights into evolving human behaviours by centralising the human body in small-group interaction-related contexts.

Thoughts & Musings

Get in touch

sanket@sabharwal.dev

Genova, Italy