Afra Feyza Akyürek

Afra Feyza Akyürek

PhD Student in Computer Science

Boston University

Hi there, this is Feyza!

I am a fourth-year Computer Science PhD student at Boston University focusing natural language processing. Before that I received my masters in Statistics from Carnegie Mellon University. I earned my bachelors both in Industrial and Computer Engineering from Koç University, Istanbul. My current research interests are on leveraging natural language for continual learning, collaborative language models, learning from natural language feedback and model editing. In the past, I worked on a diverse set of problems including resource-constrained continual learning, low-resource machine translation and ethics in NLP.

Interests

  • Natural Language Processing
  • Learning from Human and Machine Feedback
  • Collaborative Language Models

Education

  • PhD in Computer Science, 2024

    Boston University

  • MSc in Statistics, 2019

    Carnegie Mellon University

  • BSc in Computer Engineering, 2018

    Koç University

  • BSc in Industrial Engineering, 2018

    Koç University

Publications

Afra Feyza Akyürek, Ekin Akyürek, Ashwin Kalyan, Peter Clark, Derry Wijaya, Niket Tandon (2023). Generating Natural Language Feedback with Reinforcement Learning. In submission.

Lei Guo, Yiyan Zhang, Kate Mays, Afra Feyza Akyürek, Derry Wijaya, Margrit Betke (2023). Agenda Setting, Cross-cutting Effects, and Political Expression on Social Media The Gun Violence Case. Communications Research.

PDF

Afra Feyza Akyürek, Muhammed Yusuf Kocyigit, Sejin Paik, Derry Wijaya (2022). Challenges in Measuring Bias via Open-Ended Language Generation. Gender Bias in NLP Workshop at NAACL 2022.

PDF Code

Afra Feyza Akyürek, Sejin Paik, Muhammed Yusuf Kocyigit, Seda Akbiyik, Şerife Leman Runyun, Derry Wijaya (2022). On Measuring Social Biases in Prompt-Based Learning. Findings of NAACL 2022.

PDF Code Dataset

Afra Feyza Akyürek, Ekin Akyürek, Derry Wijaya, Jacob Andreas (2022). Subspace Regularizers for Few-Shot Class Incremental Learning. ICLR 2022.

PDF Code

Garry Kuwanto*, Afra Feyza Akyürek*, Isidora Chara Tourni*, Siyang Li*, Derry Wijaya (2022). Low-Resource Machine Translation Training Curriculum Fit for Low-Resource Languages. Preprint.

PDF

Ekin Akyürek, Afra Feyza Akyürek, Jacob Andreas (2021). Learning to Recombine and Resample Data for Compositional Generalization. ICLR 2021.

PDF Code

Haryo Akbarianto Wibowo, Made Nindyatama Nityasya, Afra Feyza Akyürek, Suci Fitriany, Alham Fikri Aji, Radityo Eko Prasojo, Derry Tanti Wijaya (2021). IndoCollex: A Testbed for Morphological Transformation of Indonesian Word Colloquialism. Findings of ACL 2021.

PDF Code Dataset

Afra Feyza Akyürek, Lei Guo, Randa Elanwar, Margrit Betke, Prakash Ishwar, Derry T. Wijaya (2020). Multi-Label and Multilingual News Framing Analysis. In ACL 2020.

PDF Code Dataset

Experience

 
 
 
 
 

Research Intern

Allen Institute for AI

Sep 2022 – Dec 2022 Seattle, Washington
Learning to generate natural language feedback automatically with reinforcement learning for repairing model outputs.
 
 
 
 
 

Machine Learning Research Intern

Apple

Jun 2021 – Aug 2021 Cambridge, Massachusetts
Measuring and tackling gender bias in large-scale pretained language models.
 
 
 
 
 

Biostatistics Intern

Novartis Pharmaceuticals

May 2019 – Jul 2019 East Hanover, New Jersey
Responsibilities included high-dimensional statistical analysis of clinical trial data for CAR-T therapy, a highly promising personalized immunotherapy for cancer. I created an explainable predictive modelfor treatment efficacy, revealed outstanding findings and communicated them to clinical experts.
 
 
 
 
 

Operations Research Intern

PlusOneMinusOne (Arute Solutions)

Jun 2017 – Aug 2017 Istanbul, Turkey
Implementation of meta-heuristic algorithms such as genetic algorithm and simulated annealing to address cash management across ATM networks.
 
 
 
 
 

Logistics Analytics Intern

LC Waikiki

Jun 2016 – Dec 2016 Istanbul, Turkey
Responsibilities included determining optimal load factors for trucks using CPLEX Optimizer and dispersion analysis of goods across the logistics center.

Contact