Teaching
Courses & Seminars
Teaching
I draw meaning and satisfaction out of teaching, and see it as an important role of my job in academia. I strive for effective communication, and believe that every subject can be made accessible to any audience. I regularly give meta-presentations about How to Write and How to Present, my pale spinoff of the legendary How to Speak lecture by Prof. Patrick Winston. I won several teaching awards at the Hebrew University.
Courses
Introduction to Machine Learning
An introductory course to the field of machine learning, covering the foundations of statistical learning, and the applicability of machine learning to real world problems. We focus on the PAC model, addressing fundamental questions like: What is machine learning? What type of concepts are learnable? How can we learn from data? We build a machine learning toolbox, learning the foundations and implementation of linear regression, various classifiers (SVM, decision trees, logistic regression), unsupervised learning (clustering, dimensionality reduction), ensembles, and deep learning (gradient descent and neural network architectures), as well as overarching concepts such as the bias-variance tradeoff and regularization.
Advanced Natural Language Processing
Brings students up to date with cutting edge topics in NLP research, with special emphasis on topics pursued by Roy Schwartz and me, including efficiency, multilingual NLP, evaluation and more. Students work on a semester long research projects of their choice under the supervision of the course staff.
Seminars
A Seminar on Causality
A graduate seminar on causality, reading and discussing foundational and recent work on causal reasoning, causal inference, and their connections to machine learning and NLP.
On Language Models and Consciousness: A Seminar on NLP and Philosophy
We read papers interleaving seminal work on the development of language models from Word2Vec to ChatGPT with philosophical examinations of what it means to be conscious, including Turing (1950), Nagle (1974), Searle (1980), and more. See below the complete reading list, and an ACL 2024 publication based on this seminar.
A Seminar on Multilingual NLP
We read papers about the state of multilingual NLP, in terms of tokenizers, language modeling, datasets, and more. See below an outstanding NAACL 2022 paper which begun as part of this seminar.