Anonymous Learning via Look-Alike Clustering: A Precise Analysis of Model Generalization
Authors: Adel Javanmard, Vahab Mirrokni
NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we corroborate our asymptotic theory with finite-sample numerical experiments where we observe a perfect match when the sample size is only of order of a few hundreds. |
| Researcher Affiliation | Collaboration | Adel Javanmard University of Southern California, Google Research ajavanma@usc.edu Vahab Mirrokni Google Research mirrokni@google.com |
| Pseudocode | No | The paper provides mathematical derivations and descriptions of procedures, but no formal pseudocode or algorithm blocks are included. |
| Open Source Code | No | The paper does not contain any explicit statements about making its source code open, nor does it provide links to a code repository. |
| Open Datasets | No | We consider a linear regression setting, where we are given n i.i.d pairs (xi,yi), where the response yi is given by yi = xi,θ0 + εi, εi N(0,σ2). We assume that there is a clustering structure on features xi, i [n], independent from the responses. We model this structure via Gaussian-Mixture model. |
| Dataset Splits | No | The paper mentions generating data and using a 'test set of size 50K' in Section 6, but it does not provide specific percentages or counts for training, validation, and test splits for reproducibility. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU/CPU models, memory, or cloud computing specifications used for running the experiments. |
| Software Dependencies | No | The paper does not list specific software libraries or their version numbers used in the implementation or for analysis. |
| Experiment Setup | Yes | In our experiments, we vary the sample size n and plot the risk of θL and θ versus ψd ψp = (d p)/n. We consider different settings, where we vary rs, rns and σ (noise variance in model (2.1)). In Figure 2, we report the results. Curves correspond to our asymptotic theory and dots to the numerical simulations. (Each dot is obtained by averaging over 20 realizations of that configuration.) |