Avinash Saravanan

Detroit Metropolitan Area Contact Info
219 followers 215 connections

Join to view profile

About

Experienced Full Stack Software Engineer and Computer Scientist currently based in the…

Articles by Avinash

Activity

Join now to see all activity

Experience & Education

  • REPAIRIFY, INC.

View Avinash’s full experience

See their title, tenure and more.

or

By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.

Licenses & Certifications

Volunteer Experience

Publications

  • Giving Social Robots a Conversational Memory for Motivational Experience Sharing

    IEEE

    In ongoing and consecutive conversations with persons, a social robot has to determine which aspects to remember and how to address them in the conversation. In the health domain, important aspects concern the health-related goals, the experienced progress (expressed sentiment) and the ongoing motivation to pursue them. Despite the progress in speech technology and conversational agents, most social robots lack a memory for such experience sharing. This paper presents the design and evaluation…

    In ongoing and consecutive conversations with persons, a social robot has to determine which aspects to remember and how to address them in the conversation. In the health domain, important aspects concern the health-related goals, the experienced progress (expressed sentiment) and the ongoing motivation to pursue them. Despite the progress in speech technology and conversational agents, most social robots lack a memory for such experience sharing. This paper presents the design and evaluation of a conversational memory for personalized behavior change support conversations on healthy nutrition via memory-based motivational rephrasing. The main hypothesis is that referring to previous sessions improves motivation and goal attainment, particularly when references vary. In addition, the paper explores how far motivational rephrasing affects user’s perception of the conversational agent (the virtual Furhat). An experiment with 79 participants was conducted via Zoom, consisting of three conversation sessions. The results showed a significant increase in participants’ change in motivation when multiple references to previous sessions were provided.

    Other authors
    See publication

  • Towards a Real-time Measure of the Perception of Anthropomorphism in Human-robot Interaction

    Association for Computing Machinery

    Published within MuCAI'21: Proceedings of the 2nd ACM Multimedia Workshop on Multimodal Conversational AI which is a part of ACM Multimedia.

    How human-like do conversational robots need to look to enable long-term human-robot conversation? One essential aspect of long-term interaction is a human's ability to adapt to the varying degrees of a conversational partner's engagement and emotions. Prosodically, this can be achieved through (dis)entrainment. While speech-synthesis has been a…

    Published within MuCAI'21: Proceedings of the 2nd ACM Multimedia Workshop on Multimodal Conversational AI which is a part of ACM Multimedia.

    How human-like do conversational robots need to look to enable long-term human-robot conversation? One essential aspect of long-term interaction is a human's ability to adapt to the varying degrees of a conversational partner's engagement and emotions. Prosodically, this can be achieved through (dis)entrainment. While speech-synthesis has been a limiting factor for many years, restrictions in this regard are increasingly mitigated. These advancements now emphasise the importance of studying the effect of robot embodiment on human entrainment. In this study, we conducted a between-subjects online human-robot interaction experiment in an educational use-case scenario where a tutor was either embodied through a human or a robot face. 43 English-speaking participants took part in the study for whom we analysed the degree of acoustic-prosodic entrainment to the human or robot face, respectively. We found that the degree of subjective and objective perception of anthropomorphism positively correlates with acoustic-prosodic entrainment.

    Other authors
    See publication

  • Masters Thesis: Experience Sharing Conversational Agent for Type II Diabetes

    TU Delft Education Repository

    In this work, a conversational interaction was designed and implemented to test the effect of references to past events or shared experiences rephrased into motivational phrases within the context of working towards a diet related goal that can assist with type II diabetes over multiple sessions.Prior works that utilized a memory typically did not utilize a memory to refer to the past within conversations and when past events were referred to within a conversation, they were not utilized with…

    In this work, a conversational interaction was designed and implemented to test the effect of references to past events or shared experiences rephrased into motivational phrases within the context of working towards a diet related goal that can assist with type II diabetes over multiple sessions.Prior works that utilized a memory typically did not utilize a memory to refer to the past within conversations and when past events were referred to within a conversation, they were not utilized with motivational rephrasing. Most prior works did not analyze long term interaction or how references to past events should have been utilized. To further research in this area, the following research questions were posed.
    1.) Does referring to previous sessions improve goal attainment?
    2.) Does referring to previous sessions improve user experience?
    3.) Does a variety of references differ from making the same reference?
    To determine the answers to these questions, interactions were carried out through Zoom in a between subjects experiment with three groups of participants.The results, once analyzed, found that there was a significant difference between interactions that only make references to the same shared experience and interactions that make references to various shared experiences where various shared experiences resulted in a significant increase in the change in motivation from prior to the experiment to during the experiment.This offers the following contributions:
    1.) Provide clarity on how shared experiences should be used and not only whether they should be used.
    2.) A motivational memory utilizing shared experiences with intrinsic motivational value.
    3.) Determine the effect of motivational references on goal achievement and user experience.
    This holds promise for the use of such a memory for goal based efforts and future work can further the domains of application and effectiveness of the motivational memory and shared experiences.

    See publication

Courses

  • Analytics and Machine Learning for Software Engineering

    IN4334

  • Artificial Intelligence Techniques

    IN4010-12

  • Computer Game Design and Development

    EECS 494

  • Conversational Agents

    CS4270

  • Crowd Computing

    CS4145

  • Data Structures and Algorithms

    EECS 281

  • Data Visualization

    IN4086-14

  • Database Management Systems

    EECS 484

  • Deep Learning

    CS4240

  • Dutch Elementary 1

    WM1115TU

  • Elementary Programming Concepts

    EECS 183

  • Entrepreneurship Basic Course

    MOT9610

  • Evolutionary Algorithms

    CS4205

  • Idea to Start-up – IT and AI

    TPM411A

  • Information Retrieval

    IN4325

  • Introduction to Artificial Intelligence

    EECS 492

  • Introduction to Computer Organization

    EECS 370

  • Introduction to Computer Security

    EECS 388

  • Introduction to Operating Systems

    EECS 482

  • Introduction to Statistics and Data Analysis

    Statistics 250

  • Machine Learning 1

    CS4220

  • Media Japanese I

    ASIANLAN 425

  • Multimedia Search and Recommendation

    CS4065

  • Networking

    EE4C06

  • Programming and Introductory Data Structures

    EECS 280

  • Seminar Social Signal Processing

    CS4165

  • Socio-Cognitive Engineering

    CS4235

  • Technical Writing

    WM0201TU-ENG

Projects

  • Personal Website

    - Present

    A personal website showcasing past projects, interests, etc.

    See project

  • Prosodic Feature Extraction

    -

    Applied machine learning techniques and statistics to analyze and annotate behavior within social contexts as a group. The problem focused on in this project is that of the relationship between prosody and memory. Specifically, we analyze how those interacting with a conversational agent or speaker entrain their prosody to a tutor's speech over time based on whether the participant of the experiment is interacting with a tutor with a human face or a robot face. To do this, we showed…

    Applied machine learning techniques and statistics to analyze and annotate behavior within social contexts as a group. The problem focused on in this project is that of the relationship between prosody and memory. Specifically, we analyze how those interacting with a conversational agent or speaker entrain their prosody to a tutor's speech over time based on whether the participant of the experiment is interacting with a tutor with a human face or a robot face. To do this, we showed participants of the experiment videos of a human tutor and of a robotic tutor with the same audio as the human and the human's facial gestures mapped to the robot so the robot matches the movements as well. The robot used was the Furhat robot. We used recorded audio to analyze levels of entrainment, and we used a questionnaire to analyze levels of recall and recognition. Scripts were made in Python with Furhat using Kotlin. For analysis of audio, we looked at certain features such as f0, mean and max values of pitch, HNR, and different values of shimmer and jitter. To do this, we utilized the Pydub and Parselmouth libraries in python. We then calculated convergence, proximity, and synchrony to relate the features to acoustic-prosodic entrainment. To measure statistical significance, we used t-tests and mann-whitney u tests after using z-scores for standardization.

    Other creators
    See project

  • Crowd Sourcing for Entity Resolution

    -

    Designed and implemented novel crowd sourcing tasks to solve entity resolution problems. Published batches and ran experiments on Amazon Mechanical Turk.

    Other creators
    See project

  • A reproduction attempt of “Dropout: A simple way to prevent neural networks from overfitting”

    -

    Reproduced the results of the following paper: http://jmlr.org/papers/v15/srivastava14a.html

    Other creators
    See project

  • Ad Hoc Wikipedia Table Retrieval

    -

    Reproduction of the results in this paper:https://arxiv.org/pdf/1802.06159.pdf to generate ranked lists of tables from Wikipedia (similar to a search engine). Used SPARQL to access DBPedia.

    Other creators
    See project

  • Stance Detection in Article Headlines

    -

    The goal of the project was to reproduce this paper:https://www.aclweb.org/anthology/N16-1138/ on stance classification for microblogs and news articles. The system has been trained to determine the article headline stance (for, against or observing) with respect to the claim. The team was able to reproduce the entire research and actually improve the model accuracy.

    Other creators
    See project

  • MIT 6.S094 Deep Traffic Solution

    -

    My Solution to the MIT Deep Traffic Project for MIT 6.S094. https://selfdrivingcars.mit.edu/deeptraffic/

    Visualizations can be viewed here: https://asarav.github.io/src/html/projects.html#deep-traffic

    See project

  • Rebel Knight

    -

    Final Project for EECS 494.
    An endless runner type game built in Unity Game Engine. Coded in C#.
    Can be played here: https://asarav.github.io/src/games/Rebel_Knight/Rebel_Knight.html

    Other creators
    See project

Honors & Awards

  • JNHS College Chapter Induction

    AATJ

    https://www.aatj.org/resources/studentactivities/jnhs/JNHSCollege_List_2016.pdf
    The Japanese National Honor Society – College Chapter (JNHS–CC) recognizes and encourages scholastic achievement and excellence in the study of the Japanese language.

  • William J. Branstrom Freshman Prize

    University of Michigan

    First-term freshmen who rank in the upper five percent of their class within their school or college are awarded the William J. Branstrom Freshman Prize. The student must have taken at least 14 graded (A-E) credits during the fall 2012 term to be eligible for this award. Advanced placement credit does not disqualify a student for consideration of this award.

Test Scores

  • JLPT N2

    Score: Passed

    Passed N2 level of the Japanese language proficiency test and received certification.

  • JLPT N4

    Score: Passed

    Passed N4 level of the Japanese language proficiency test and received certification.

Languages

  • Japanese

    Full professional proficiency

  • English

    Native or bilingual proficiency

  • Dutch

    Elementary proficiency

  • Korean

    Elementary proficiency

More activity by Avinash

View Avinash’s full profile

  • See who you know in common
  • Get introduced
  • Contact Avinash directly
Join to view full profile

Other similar profiles

Explore more posts

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Others named Avinash Saravanan

11 others named Avinash Saravanan are on LinkedIn

See others named Avinash Saravanan

Add new skills with these courses

See all courses