Tejas Gokhale

Lab Mission

Our lab mission is to conduct innovative research towards the development of computational theories of visual perception and its links to action, learning, reasoning, and communication.

Lab Values

We are a group that operates on two primary values: RESPECT and RIGOR.
  • Respect: We help and support each other to do the best work that we can. This includes participating in discussions, asking questions, giving feedback on papers and presentations, debugging code, celebrating success and navigating failure. We express our opinions respectfully and in good faith. There is no hierarchy in the group. Tejas isn't your boss -- he is a friend, advisor, and facilitator in your scientific journey. You shouldn't call him Professor/Dr/Sir/Lord/Supreme Leader etc.
  • Rigor: The work that we do is rigorous. We are grounded in fundamentals. We pay attention to detail. We are proud of what we do, we are always looking to improve our skills, and we engage with feedback and constructive criticism. We don't gate-keep. We are transparent. We accept our mistakes.

Lab Membership and Preparation

  • Joining the Group: If you are interested in working with me, please fill out this form. Your answers to questions in the form will help us prepare for a discussion if I reach out for a short interview. If you are at UMBC, you can also visit me during my office hours. The best way to get a headstart into my research area would be to take one of my classes.
    • PhD students: I recruit PhD students every Fall.
    • MS students: I only recruit a small number of MS thesis students who take one of my classes. I don't offer short-term RA or volunteer positions.
    • Undergraduate students: Reach out if you're passionate about machine learning, computer vision, NLP, or related areas.

  • Classes to take: I recommend taking the following UMBC graduate classes (or their equivalent elsewhere) for building a foundation for graduate research with us:
    • CMSC 691 Computer Vision [required]
    • CMSC 678 Machine Learning [required]
    • CMSC 673 Natural Language Processing [required]
    • CMSC 679 Robotics
    • CMSC 634 Computer Graphics
    • CMSC 671 Artificial Intelligence
    • CMSC 675 Neural Networks / Deep Learning
    • ENEE 612 Digital Image Processing (if you haven't taken an undergrad signal/image processing class)
    • ENEE 620 Probability and Random Processes
    • ENEE 621 Detection and Estimation Theory
    • ENEE 712 Pattern Recognition

    For Ph.D. students: It is unlikely that you will contribute to philosophy (that's what the Ph. in Ph.D. stands for) if you only take classes in Computer Science -- in fact more often than not it leads to a very myopic view of the world (eg. Bay Area techbros). I encourage taking some classes outside the department (eg. genetics, animal behavior, psychology, economics, statistics, literature) and reading broadly in your free time.

  • Compensation: While I would like to work with many exceptional students, I may not always be able to accomodate everyone. Availability of funding is often a major factor that goes into such decisions. However there are several useful resources and programs at UMBC:


Being exposed to different styles of doing research, different organizations, and a wider network of collaborators, is instrumental in intellectual growth. One of the best ways of doing this is to go for internships in organizations outside the university. Internships are also a way for securing additional (and significantly higher) funding. PhD researchers in my lab will be encouraged to apply and compete for research internships and I will be extremely supportive of such endeavors. My only expectation is that the internship should make a positive contribution towards completion of your PhD (usually through a collaborative publication with me and your internship mentors). BS and MS researchers will be encouraged to apply for relevant internship and co-op opportunities that contribute to their career growth in industry.

Note for Recruiters: If you are hiring interns into your organization, please reach out to me; we can schedule a time to talk and identify suitable students and synergistic research topics.

Recommendation Letters

Letters of recommendation are an important aspect of hiring in both academia and industry -- business is inherently about people and trust. I will be glad to write you a letter if you took my class and performed well, or if you are a researcher or collaborator with my lab. Letters will be confidential. I will never write negative, average, or weak letters -- if I do not have enough evidence to write a strong letter for you, I will ask you to request a letter from someone who knows you better and can make a stronger case for you.