Pinar Ozisik


Research Scientist
Algorithmic Alignment Group
MIT CSAIL
Contact:
a.pinar.ozisik@gmail.com

About

I am a research scientist in the Algorithmic Alignment Group at MIT CSAIL, working with Prof. Dylan Hadfield-Menell. I am broadly interested in ensuring that algorithms and systems behave correctly, safely, and in line with their intended purpose in the real world. Before joining CSAIL, I was a visiting researcher in the Camera Culture group, working with Prof. Ramesh Raskar at the MIT Media Lab.

I am also on the lookout for positions in industry or academia. Please feel free to email me if you think I'd be a good fit for your group. Here is my CV and research statement.

I received my Ph.D. from the University of Massachusetts Amherst, where I was working with Prof. Brian Levine in the Cryptoeconomics Lab and collaborating with Prof. Philip Thomas in the Autonomous Learning Lab. My thesis focused on the analysis and application of concentration inequalities to preserve the desirable properties of systems. More specifically, I worked on a protocol that used tail inequalities to improve the network performance of blockchain systems; and I analyzed the susceptibility of a class of reinforcement learning algorithms that leveraged tail inequalities to data poisoning attacks.


News

Jun. 2021 I gave a talk at the Facebook Novi System Research Seminar.
Feb. 2021 I officially graduated! My thesis, titled Concentration Inequalities in the Wild: Case Studies in Blockchain & Reinforcement Learning, is online.
Aug. 2020 I defended my thesis!
Sep. 2020 Our new paper, titled Security Analysis of Safe and Seldonian Reinforcement Learning Algorithms, is accepted to NeurIPS 2020!
Sep. 2019 This semester I'm teaching a new freshman seminar titled Computer Science Brain Teasers.
Jun. 2019 Our new paper on Graphene is accepted to SIGCOMM 2019! You can take a look here.
Sep. 2018 Check out the freshman seminar I'm teaching this semester titled Ethical Issues in Technology.
Nov. 2017 Check out the discussion about our new block propagation protocol, Graphene. Bitcoin Unlimited even mentioned that they might implement it here!
Jan. 2017 You can find our tutorial explaining the derivation of Nakamato's double-spend attack here.
Jun. 2016 I'm off to New Mexico to attend the Complex Systems Summer School in the Santa Fe Institute!
May 2016 I passed the qualifying exam and now am a Ph.D. Candidate!
May 2015 I'm going to intern at a defense company, Systems and Technology Research, as part of the Analysis and Decision Systems team.
Apr. 2013 I received the Anita Borg Memorial Scholarship (now called Women Techmakers Scholarship). Thank you very much Google!

Education

Academic Experience

Industry Experience



Last modified: 03/30/2023