Hi! I'm Mikayla Timm.

Hi! I'm a 3rd year MS/PhD student in Computer Science at the University of Massachusetts Amherst researching in the Computer Vision Laboratory, where I'm advised by Subhransu Maji. I obtained my B.S. in Computer Science with a minor in Mathematics from the University of West Florida, where I graduated Summa Cum Laude in spring 2017.

My current research focuses on the intersection of computer vision, deep learning, and natural language.

This summer (2019), I'm interning at Pinterest in San Francisco, working on the Visual Search Team!

Aside from research and taking classes, I enjoy playing my flute, taking care of my cat, eating food, and going to rock concerts.

Research Experience

Machine Learning Research Intern - Pinterest (Summer 2019-Present)

  • Pinterest Labs summer intern, working on the Visual Search team.
  • Trained and evaluated new and existing vision models for fashion image retrieval through image embeddings conditioned on specific attributes (color, pattern, fabric, gender).
  • Implemented and pushed a data pipeline to production for processing and visualizing new data.
  • Utilized AWS EC2 P3 instances to train distributed deep models in the cloud on multiple GPUs.

Graduate Research Assistant - University of Massachusetts Amherst (Fall 2017-Present)

  • Researching computer vision approaches for fine-grained recognition and vision + language.
  • Investigating deep learning models for classifying attributes of textures, generating natural language descriptions of texture images, and image retrieval from natural language. For more information, click here!
  • Collected and refined two novel vision and language datasets using Amazon Mechanical Turk.
  • Investigated techniques for fine-grained visual categorization of animals at the species level and individual level. Presented a poster and published a short paper in the 2018 CVPR FGVC and WiCV workshops - check out the poster here!

Summer Research Intern - MIT Lincoln Laboratory (Summer 2017)

  • Researched natural language processing techniques for generating word embeddings in the multilingual context to enable performing NLP tasks on inherently multilingual data, such as tweets.
  • Developed a system for processing multilingual text corpora, training word embeddings, visualizing the resulting high dimensional vectors in a 2D space, and performing intrinsic evaluations on the embeddings, such as analogies and nearest neighbors.

Wearable Device Security Research Assistant - University of West Florida (Fall 2016-Spring 2017)

  • Utilized supervised learning algorithms to classify biometric data obtained from wearable device cyber attacks.
  • Published findings (second author) in a peer-reviewed conference: Reichherzer, T., Timm, M., Earley, N., Reyes, N., & Kumar, V. (2017). Using machine learning techniques to track individuals & their fitness activities. In A. Bossard, G. Lee, & L. Miller (Eds.), Proceedings of 32nd International Conference on Computers and Their Applications, March, 20-22, 2017, Honolulu, Hawaii, USA (pp. 119–124). Winona, MN, USA: ISCA.

Summer Undergraduate Research Scholar - University of West Florida (Summer 2016)

  • Researched machine learning techniques for predicting outcomes of animals in shelters.
  • Published findings (first author) in a peer-reviewed conference: Timm, M., El-Sheikh, E. (2017). An Evaluation of Machine Learning Algorithms for Prediction of Shelter Animal Outcomes. In A. Bossard, G. Lee, & L. Miller (Eds.), Proceedings of 32nd International Conference on Computers and Their Applications, March, 20-22, 2017, Honolulu, Hawaii, USA (pp. 119–124). Winona, MN, USA: ISCA.

REU Student Researcher - UMass Amherst (Summer 2015)

  • Researched computer vision algorithms for identifying individual jaguars in images to assist ecologists with their conservation efforts.
  • Presented research in both oral and poster form at UMass and UWF Student Scholars Symposium.
  • Awarded Best Student Project in Computer Science at 2016 UWF Student Scholars Symposium.

Contact Info

Thank you for viewing my website! If you'd like to get in touch with me, please send me an email at mtimm100@gmail.com.

You can find me on LinkedIn here.