Gissella Bejarano (PhD, started 2017)
Yue Zhang (PhD, started 2017)
Arpita Chakraborty (PhD, started 2017)
Alexander Van Roijen
I am looking for motivated Masters and PhD students with interest in machine learning, data mining, and natural language processing. Please send me an email with your resume if you are interested in working with me.
I try my best to reply to student emails about working with me. But sometimes I do tend to miss a few here and there. So I decided to give an overview of expectations and the kind of work in my group. Please read this before emailing me as that tells me that you are genuinely interested in working with me (and might save us both time)! :)
My work is broadly in machine learning, more specifically in probabilistic graphical models, natural language processing, and data analytics and mining. My work mostly revolves around designing and building models for interesting problems in real-world situations. I tend to draw motivation for my research from real-world scenarios. I identify problems of interest in these domains and design and build machine learning models for these problems.
PhD is an exciting journey into the world of research. The road to a successful PhD can sometimes be rough, but the process is both enlightening and fulfilling. PhD is perhaps the only time in your life when you spend thinking exlusively on a research problem and contributing to the research world.
The internet is full of mixed (and confusing!) opinions on whether you should/should not do a PhD! As someone who pondered that question for a while, before finally deciding to do a PhD (and absolutely enjoyed it!!!), here are some reasons that helped me in my decision.
Should I do a PhD? Top 5 reasons a PhD is a good idea
Ten good reasons for doing a PhD
That said, there are times when research is frustatingly slow and tiring. Here are a few tips for a succesful PhD, some especially for succeeding in my group.
1. PhD is a long process.
Research tends to be a slow and challenging process often requiring long hours. The process of developing something novel is a time-taking process requiring tremendous amount of patience, perseverance, and tenacity. But it is also extremely fun, rewarding, and exciting!!
There are numerous resources on the web on how to do a successful PhD. Here are a few I have read myself:
GRADUATE STUDY IN THE COMPUTER AND MATHEMATICAL SCIENCES: A SURVIVAL MANUAL
HOW TO SUCCEED IN GRADUATE SCHOOL: A GUIDE FOR STUDENTS AND ADVISORS
THE Ph.D. GRIND
2. Working with real data is exciting (but often hard!).
Building models that work in the real world often turns out to be a challenging task, but can be really fun if you are interested in that! Often, all our intuitions and theories break down when working with real data. Real world data tends to be extremely messy, disorganized, and unstructured. Building meaningful models often involves hours of munging data, patiently repeating experiments in the hope that the models will outperform the baselines.
3. The three Cs of graduate school: Communication, Commitment, and Coordination.
The most important skill for succeeding in graduate school is communication: to read, and to write. Understanding existing research involves reading and understanding existing scientific literature and identifying interesting research problems. Research does not just involve producing the results but also being able to communicate precisely and concisely in a way the scientific community understands.
PhD also involves tremendous amount of commitment and dedication, which involve spending long working hours, working weekends and holidays, especially around paper deadlines.
The third and most important skill required in graduate school is coordination and planning. Managing time efficiently can help you not get overwhelmed by the amount of work and allow you to make progress. Having an everyday schedule instead of working in bursts (also referred to as the daily grind!) helps you to be prepared for deadlines, and ensures a smoother experience. Avoiding procrastination and giving yourself ample buffers help you meet tough deadlines.
If you reach here, it probably means you are very interested in pursuing a PhD; please shoot me an email after applying to Binghamton. I usually fund students only after they have either taken a course with me or worked with me on research projects. I am happy to schedule a meeting with students already admitted to the program and interested in working with me. If you are a prospective student, please refer to the department webpage for information regarding applications. At the moment, I am not offering any summer internships.