Awni Altabaa

prof_pic.jpg

Kline Tower, Office 1117

219 Prospect St

New Haven, CT 06511

Hi! Welcome to my homepage.

My name is Awni. I am a PhD student in the Department of Statistics & Data Science at Yale University studying the foundations of machine learning. My wonderful advisor is Prof. John Lafferty.

My research interests lie broadly in the intersection of machine learning, statistics, and computer science. More specifically, my research aims to study questions of the following flavor:

  • What are the architectural mechanisms and inductive biases necessary for efficient learning and strong generalization in different domains?
  • What are the fundamental theoretical limits of what is or is not possible to learn under different learning paradigms?
  • To what degree can neural networks learn functions and algorithms that can generalize compositionally to out-of-distribution inputs?

Our work tackles these questions through complementary empirical investigation and theoretical analysis. My current research focus is on algorithmic generalization and reasoning in machine learning models.

selected publications

  1. relattn.png
    Disentangling and Integrating Relational and Sensory Information in Transformer Architectures
    Awni Altabaa, and John Lafferty
    Under Review, 2024
  2. post.png
    On the Role of Information Structure in Reinforcement Learning for Partially-Observable Sequential Teams and Games
    Awni Altabaa, and Zhuoran Yang
    Neural Information Processing Systems (NeurIPS), 2024
  3. ML
    Approximation of Relation Functions and Attention Mechanisms
    Awni Altabaa, and John Lafferty
    Under Review, 2024
  4. relconv.png
    Learning Hierarchical Relational Representations through Relational Convolutions
    Awni Altabaa, and John Lafferty
    Transactions on Machine Learning Research (TMLR), 2024
  5. relbottleneck.jpg
    The Relational Bottleneck as an Inductive Bias for Efficient Abstraction
    Taylor W. Webb, Steven M. Frankland, Awni Altabaa, Kamesh Krishnamurthy, and 5 more authors
    Trends in Cognitive Science (TICS), 2024
  6. rca.png
    Abstractors and Relational Cross-Attention: An Inductive Bias for Explicit Relational Reasoning in Transformers
    Awni Altabaa, Taylor Webb, Jonathan Cohen, and John Lafferty
    International Conference on Learning Representations (ICLR), 2024
  7. decrl.gif
    Decentralized Multi-Agent Reinforcement Learning for Continuous-Space Stochastic Games
    Awni Altabaa, Bora Yongacoglu, and Serdar Yüksel
    2023 IEEE American Control Conference (ACC), Mar 2023