Research map / Optimization / Optimizers
Semiparametric Nonlinear Bipartite Graph Representation Learning with Provable Guarantees
Sen Na, Yuwei Luo, Zhuoran Yang, Zhaoran Wang, Mladen Kolar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use the NSMC pseudo-likelihood instead of the Gaussian squared-loss NIMC objective | nonlinear inductive matrix completion · Gaussian model with misspecification (τ = 0.2 and 0.4) | relative test error (Θ): 0.3633 -> 0.0427; 0.6377 -> 0.0468 | Improved |
| Use the NSMC pseudo-likelihood instead of NIMC | nonlinear inductive matrix completion · Binomial model, N_B = 100, 200, and 500 | relative test error (Θ): 0.9364 -> 0.0464; 0.9248 -> 0.0441; 0.9259 -> 0.0381 | Improved |
| Use the NSMC pseudo-likelihood instead of NIMC | nonlinear inductive matrix completion · Poisson model, ReLU + ReLU and ReLU + tanh | relative test error (Θ): 10.7018 -> 0.0975; 0.5595 -> 0.0727 | Improved |
| Use the NSMC pseudo-likelihood instead of IMC | linear inductive matrix completion · Poisson model, ReLU + sigmoid | relative test error (Θ): 1.0922 -> 0.0631 | Improved |
| Use the NSMC pseudo-likelihood instead of NIMC | nonlinear inductive matrix completion · Synthetic embedding clustering, x and z | clustering error: 0.0596 -> 0.0196; 0.1725 -> 0.0147 | Improved |
| Use the NSMC pseudo-likelihood instead of NIMC | nonlinear inductive matrix completion · Segment and Covtype | clustering error: 0.0971 -> 0.0427; 0.1931 -> 0.1373 | Improved |
| Use NSMC instead of NIMC | nonlinear inductive matrix completion · Mushroom | clustering error: 0 -> 0 | No clear effect |
| Add nonlinear activation functions to the semiparametric matrix completion model (ablation) | SMC · Gaussian model, τ = 0.2 and 0.4 | relative test error (Θ): 0.9938 -> 0.0427; 0.9849 -> 0.0468 | Improved |
| Add nonlinear activation functions to the semiparametric matrix completion model (ablation) | SMC · Binomial model, N_B = 100 | relative test error (Θ): 0.9956 -> 0.0464 | Improved |
| Add nonlinear activation functions to the semiparametric matrix completion model (ablation) | SMC · Poisson model, ReLU + ReLU | relative test error (Θ): 0.5559 -> 0.0975 | Improved |
Closest papers by meaning
- Learning Bipartite Graphs: Heavy Tails and Multiple Components (NeurIPS 2022 · similarity 0.88)
- Graph Structure Inference with BAM: Neural Dependency Processing via Bilinear Attention (NeurIPS 2024 · similarity 0.86)
- On Semiparametric Exponential Family Graphical Models (JMLR 2018 · similarity 0.86)
- Sterling: Synergistic Representation Learning on Bipartite Graphs (AAAI 2024 · similarity 0.85)
- SDMG: Smoothing Your Diffusion Models for Powerful Graph Representation Learning (ICML 2025 · similarity 0.85)
- Learning Graph Representation via Graph Entropy Maximization (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
