Tasfia Mashiat

tmashiat@gmu.edu

Achievements

  • Outstanding Graduate Teaching Assistant, Department of Computer Science, George Mason University (2021).
  • Recipient of the Computer Science Department’s Ph.D. Research Initiation Awards, George Mason University (2020)
  • Champion, Women Innovation Camp-2016, Arranged by A2I, ICT Division, Prime Minister’s Office, Bangladesh (2016).
  • Dean’s Award for Excellence in Academic Results in session 2015-2016. Khulna University of Engineering & Technology (KUET).

Hi, I am a Ph.D. Candidate in George Mason University. My advisor is Prof. Sanmay Das.

I completed my Bachelor of Science (B.Sc.) in Computer Science and Engineering from Khulna University of Engineering & Technology, Khulna, Bangladesh.

Recent News!

  • February, 2023: Will participate in Doctoral Consortium at AAMAS 2023!
  • January, 2023: Paper accepted as Extended Abstract in AAMAS 2023!
  • October, 2022: Participated in Doctoral Consortium at ACM EAAMO 2022!
  • October, 2022: Served as a local chair in ACM EAAMO 2022!
  • June, 2022: Served as a session chair in ACM FAccT 2022!
  • April, 2022: Paper Accepted in ACM FAccT 2022!
  • March, 2022: I will work as an Applied Scientist Intern at ML Solutions Lab, Amazon!

Publications

  • Tasfia Mashiat, Alex DiChristofano, Patrick J. Fowler, Sanmay Das, Beyond Eviction Prediction: Leveraging Local Spatiotemporal Public Records to Inform Action. (ACM FAccT 2024) . Paper.

  • Md Naimul Hoque, Tasfia Mashiat, Bhavya Ghai, Cecilia Shelton, Fanny Chevalier, Kari Kraus, Niklas Elmqvist. The HaLLMark Effect: Supporting Provenance and Transparent Use of Large Language Models in Writing with Interactive Visualization. (ACM CHI 2024). Paper

  • Tasfia Mashiat, Xavier Gitiaux, Huzefa Rangwala, Sanmay Das, Counterfactually Fair Dy- namic Assignment: A Case Study on Policing.(AAMAS 2023) Extended Abstract, Poster

  • Tasfia Mashiat, Xavier Gitiaux, Huzefa Rangwala, Patrick J. Fowler, Sanmay Das.Trade-offs between Group Fairness Metrics in Societal Resource Allocation.(ACM FAccT 2022) . Paper.

  • Tasfia Mashiat, Xavier Gitiaux, Huzefa Rangwala, Sanmay Das. Fairness-Aware Resource Assignment: A Case Study on Policing.The 15th Workshop for Women in Machine Learning (WiML), Neural Information Processing Systems (NeurIPS 2020) . Poster

  • Nur Imtiazul Haque, Kazi Md. Rokibul Alam, Tasfia Mashiat,Yasuhiko Morimoto. A Technique to Enrich the Secrecy Level of High Capacity Data Hiding Steganography Technique in JPEG Compressed Image. International Conference on Networking, Systems and Security (NSysS 2018), Dhaka, Bangladesh. Paper

Projects

  • Equitable Resource Allocation in the Context of Homelessness
    In the project, we aim to propose a statistically efficient approach to ensure equitable allocation of homelessness services within different subgroups of population based on protected attributes such as age, gender, race, etc. This is an on-going project.

  • Fair Assignment of Policing Resources
    We proposed a causality-based approach for the allocation of limited policing resources within neighborhoods while ensuring the allocation is fair. We conducted experiments on both synthetic and real-world data. We showed that following our approach both over-policing and under-policing can be reduced in areas with a significant difference in population demographic.

  • A Deep Neural Network Approach for Retrofitting Word Embeddings
    In this project, we proposed a deep learning model that can identify morphologically related word embeddings for languages with a high morpheme-per-word ratio in a sparse word vector domain. This project was a part of the CS747: Deep Learning Course.

  • Examining The 4G/5G LTE Cellular Network Infrastructure(Advisor: Prof. Duminda Wijesekera)
    The goal of this project was to set up 4G LTE and 5G cellular network testbeds to examine the data communication between User equipments and Base stations with the core network. This project was conducted as a part of CS701: Research Experience in CS.

  • Examining the Patterns of Taxi-rides in NYC
    The main goal of this project was to understand the demand pattern of taxi service in NYC through exploratory data analysis and machine intelligence. This project was conducted as a part of CS700: Research in CS.