Research map / Architectures and efficiency / RNNs and normalization
Scalable Monotonic Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a scalable neural network architecture for learning models that are guaranteed monotonic with respect to a subset of inputs, addressing limitations of prior monotonic methods.
Problem
Learning neural networks whose outputs are monotonic with respect to a subset of inputs, a requirement in many practical domains; existing methods may not guarantee monotonicity, need extra inference time, or lack scalability.
Approach
A monotonic neural network design intended to scale while guaranteeing monotonicity with respect to chosen inputs, avoiding the drawbacks of prior approaches (per the abstract; details not available in the provided text).
Closest papers by meaning
- Certified Monotonic Neural Networks (NeurIPS 2020 · similarity 0.90)
- Smooth Min-Max Monotonic Networks (ICML 2024 · similarity 0.89)
- Advancing Constrained Monotonic Neural Networks: Achieving Universal Approximation Beyond Bounded Activations (ICML 2025 · similarity 0.89)
- Constrained Monotonic Neural Networks (ICML 2023 · similarity 0.87)
- Scalable Neural Network Kernels (ICLR 2024 · similarity 0.85)
- Learning Monotonic Probabilities with a Generative Cost Model (ICML 2025 · 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.