Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Spaces:
DineshAI
/
uG4IOdaAGk
Running

App Files Files Community
uG4IOdaAGk / repro /src
Ctrl+K
Ctrl+K
  • 1 contributor
History: 3 commits
DineshAI's picture
DineshAI
Reproduce claim 4 on REAL Census/Adult: 45222 samples, MLP 81.5%, DAG alters attribution (0.048), priority sweep separates sensitive (capital-gain) from stable (race) features
e926cb3 verified 11 days ago
  • verify_census_pasv.py
    8.21 kB
    Reproduce claim 4 on REAL Census/Adult: 45222 samples, MLP 81.5%, DAG alters attribution (0.048), priority sweep separates sensitive (capital-gain) from stable (race) features 11 days ago
  • verify_mnist_pasv.py
    7.59 kB
    Reproduce claim 3 on REAL MNIST-800: 8 providers/DAG/10-NN, PASV penalizes Copier (0.026 vs Owner 0.492) and Poisoner (negative, decreasing in b), Boosters stable 11 days ago
  • verify_pasv.py
    10.3 kB
    Reproduce Thm 3.12 PASV axioms (Efficiency/Null/Linearity/EWU==Shapley/Monotonicity exact) + data-valuation mechanism: DAG penalizes copier (0.008 vs Shapley 0.059) and poisoner (negative) 11 days ago