Artificial Intelligence and Information Systems Lab

 

Artificial Intelligence and Information Systems Laboratory (AI-ISL) is a deep learning research lab, focuses on information theoretic approach on various machine learning problems.

 

 

News

  • We are looking for passionate new graduate students and undergrad interns to join our team.
     - Check Join Us
  • Our group has 1 paper accepted to ICML 2024
     - Improved Communication-Privacy Trade-offs in L2 Mean Estimation under Streaming Differential Privacy
  • ISL moved to department of Artificial Intelligence at Yonsei University.
  • Our group has 1 papers accepted to NeurIPS 2023 Workshop on Diffusion Models
     - LoRA can Replace Time and Class Embeddings in Diffusion Probabilistic Models
  • Our group has 2 papers accepted to NeurIPS 2023
     - Censored Sampling of Diffusion Models Using 3 Minutes of Human Feedback
     - Exact Optimality of Communication-Privacy-Utility Tradeoffs in Distributed Mean Estimation
  • Our group has 1 papers accepted to ICCV 2023 Workshop on Low-Bit Quantized Neural Networks
     - Fully Quantized Always-on Face Detector Considering Mobile Image Sensors
  • Our group has 1 papers accepted to ICML 2023 Workshop on Federated Learning
     - Exact Optimality of Communication-Privacy-Utility Tradeoffs in Distributed Mean Estimation
  • Our group has 1 paper accepted to ECCV 2022
     - Prune Your Model Before Distill It
  • Our group has 1 paper accepted to ICML 2022
     - Neural Tangent Kernel Analysis of Deep Narrow Neural Networks
  • Our group has 1 paper accepted to AISTATS 2022
     - An Information-Theoretic Justification for Model Pruning
  • Our group has 1 paper accepted to ICML 2021
     - WGAN with an Infinitely Wide Generator Has No Spurious Stationary Points