Lab for Interpretability and Theory-Driven Deep Learning
Lab for Interpretability and Theory-Driven Deep Learning, SJTU
China
Pinned Repositories
aog
PyTorch Implementation of the paper "Defining and Quantifying the Emergence of Sparse Concepts in DNNs" (CVPR 2023)
BNN-concepts
PyTorch implementation of the paper "Bayesian Neural Networks Avoid Encoding Complex and Perturbation-Sensitive Concepts" (ICML 2023)
generalizable-interaction
PyTorch implementation of the paper "Defining and extracting generalizable interaction primitives from DNNs" (ICLR 2024)
interaction-concept
PyTorch implementation of the paper "Does a Neural Network Really Encode Symbolic Concept?" (ICML 2023)
interaction-sparsity
PyTorch implementation of the paper "Where We Have Arrived in Proving the Emergence of Sparse Interaction Primitives in AI Models" (ICLR 2024)
InteractionDynamics
Academic page for the empirical and theoretical findings of the two-phase dynamics of interactions
ReasoningMemorization
Academic page for Paper: Quantifying In-Context Reasoning Effects and Memorization Effects in LLMs
sjtu-xai-lab.github.io
Quanshi Zhang's website. Sjtu interpretable ml lab website.
transformation-complexity
PyTorch implementation of "Towards Theoretical Analysis of Transformation Complexity of ReLU DNNs"
UnifyAttribution
Lab for Interpretability and Theory-Driven Deep Learning's Repositories
sjtu-xai-lab/interaction-concept
PyTorch implementation of the paper "Does a Neural Network Really Encode Symbolic Concept?" (ICML 2023)
sjtu-xai-lab/aog
PyTorch Implementation of the paper "Defining and Quantifying the Emergence of Sparse Concepts in DNNs" (CVPR 2023)
sjtu-xai-lab/generalizable-interaction
PyTorch implementation of the paper "Defining and extracting generalizable interaction primitives from DNNs" (ICLR 2024)
sjtu-xai-lab/sjtu-xai-lab.github.io
Quanshi Zhang's website. Sjtu interpretable ml lab website.
sjtu-xai-lab/BNN-concepts
PyTorch implementation of the paper "Bayesian Neural Networks Avoid Encoding Complex and Perturbation-Sensitive Concepts" (ICML 2023)
sjtu-xai-lab/Learn
PyTorch implementation of the paper "Towards the Difficulty for a Deep Neural Network to Learn Concepts of Different Complexities" in (NeurIPS 2023)
sjtu-xai-lab/transformation-complexity
PyTorch implementation of "Towards Theoretical Analysis of Transformation Complexity of ReLU DNNs"
sjtu-xai-lab/interaction-sparsity
PyTorch implementation of the paper "Where We Have Arrived in Proving the Emergence of Sparse Interaction Primitives in AI Models" (ICLR 2024)
sjtu-xai-lab/InteractionDynamics
Academic page for the empirical and theoretical findings of the two-phase dynamics of interactions
sjtu-xai-lab/ReasoningMemorization
Academic page for Paper: Quantifying In-Context Reasoning Effects and Memorization Effects in LLMs
sjtu-xai-lab/UnifyAttribution
sjtu-xai-lab/InteractionSparsity