Copertina

Activation Bottleneck: Sigmoidal Neural Networks Cannot Forecast a Straight Line

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Toller, Maximilian, Hussain, Hussain, Geiger, Bernhard C
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Constraining Anomaly Detection with Anomaly-Free Regions
di: Toller, Maximilian, et al.
Pubblicazione: (2024)

Information Plane Analysis of Binary Neural Networks
di: Nothnagel, Maximilian, et al.
Pubblicazione: (2026)

Trustworthy Representation Learning via Information Funnels and Bottlenecks
di: de Freitas, João Machado, et al.
Pubblicazione: (2022)

Taming the Sigmoid Bottleneck: Provably Argmaxable Sparse Multi-Label Classification
di: Grivas, Andreas, et al.
Pubblicazione: (2023)

An Eulerian Perspective on Straight-Line Sampling
di: Tsimpos, Panos, et al.
Pubblicazione: (2025)

Approximating Families of Sharp Solutions to Fisher's Equation with Physics-Informed Neural Networks
di: Rohrhofer, Franz M., et al.
Pubblicazione: (2024)

Data vs. Physics: The Apparent Pareto Front of Physics-Informed Neural Networks
di: Rohrhofer, Franz M., et al.
Pubblicazione: (2021)

On the Role of Priors in Bayesian Causal Learning
di: Geiger, Bernhard C., et al.
Pubblicazione: (2025)

Graph-Based Neural Models for Transonic Aerodynamics: AeroFormer & MeshGAT Architectures
di: Hussain, Saad
Pubblicazione: (2025)

Straight-Line Diffusion Model for Efficient 3D Molecular Generation
di: Ni, Yuyan, et al.
Pubblicazione: (2025)

An Invitation to Deep Reinforcement Learning
di: Jaeger, Bernhard, et al.
Pubblicazione: (2023)

Automated Design of Linear Bounding Functions for Sigmoidal Nonlinearities in Neural Networks
di: König, Matthias, et al.
Pubblicazione: (2024)

Horizon Activation Mapping for Neural Networks in Time Series Forecasting
di: Hans, Krupakar, et al.
Pubblicazione: (2026)

Beyond Gaussian Initializations: Signal Preserving Weight Initialization for Odd-Sigmoid Activations
di: Lee, Hyunwoo, et al.
Pubblicazione: (2025)

Universal Outlier Hypothesis Testing via Mean- and Median-Based Tests
di: Geiger, Bernhard C., et al.
Pubblicazione: (2026)

Discovering the Representation Bottleneck of Graph Neural Networks
di: Wu, Fang, et al.
Pubblicazione: (2022)

B-PL-PINN: Stabilizing PINN Training with Bayesian Pseudo Labeling
di: Innerebner, Kevin, et al.
Pubblicazione: (2025)

Stabilizing PINNs: A regularization scheme for PINN training to avoid unstable fixed points of dynamical systems
di: Babic, Milos, et al.
Pubblicazione: (2025)

Extending Straight-Through Estimation for Robust Neural Networks on Analog CIM Hardware
di: Feng, Yuannuo, et al.
Pubblicazione: (2025)

Efficient Logistic Regression with Mixture of Sigmoids
di: Di Gennaro, Federico, et al.
Pubblicazione: (2026)

Why Cannot Neural Networks Master Extrapolation? Insights from Physical Laws
di: Dakhmouche, Ramzi, et al.
Pubblicazione: (2025)

Beyond Softmax: Dual-Branch Sigmoid Architecture for Accurate Class Activation Maps
di: Oh, Yoojin, et al.
Pubblicazione: (2025)

On the Convergence and Straightness of Rectified Flow
di: Bansal, Vansh, et al.
Pubblicazione: (2024)

Making Sigmoid-MSE Great Again: Output Reset Challenges Softmax Cross-Entropy in Neural Network Classification
di: Tyagi, Kanishka, et al.
Pubblicazione: (2024)

Combining Causal Models for More Accurate Abstractions of Neural Networks
di: Pîslar, Theodora-Mara, et al.
Pubblicazione: (2025)

Analysis of Using Sigmoid Loss for Contrastive Learning
di: Lee, Chungpa, et al.
Pubblicazione: (2024)

Semiring Activation in Neural Networks
di: Smets, Bart M. N., et al.
Pubblicazione: (2024)

Deep Neural Network Initialization with Sparsity Inducing Activations
di: Price, Ilan, et al.
Pubblicazione: (2024)

Novel Deep Learning Architecture for Heart Disease Prediction using Convolutional Neural Network
di: Hussain, Shadab, et al.
Pubblicazione: (2021)

StraightLine: An End-to-End Resource-Aware Scheduler for Machine Learning Application Requests
di: Ching, Cheng-Wei, et al.
Pubblicazione: (2024)

The Interaction Bottleneck of Deep Neural Networks: Discovery, Proof, and Modulation
di: Deng, Huiqi, et al.
Pubblicazione: (2025)

Topological Neural Networks: Mitigating the Bottlenecks of Graph Neural Networks via Higher-Order Interactions
di: Giusti, Lorenzo
Pubblicazione: (2024)

Optimizer Dynamics at the Edge of Stability with Differential Privacy
di: Hussain, Ayana, et al.
Pubblicazione: (2025)

Setting the Record Straight on Transformer Oversmoothing
di: Dovonon, Gbètondji J-S, et al.
Pubblicazione: (2024)

Robust Filtering -- Novel Statistical Learning and Inference Algorithms with Applications
di: Chughtai, Aamir Hussain
Pubblicazione: (2025)

Global Minimizers of Sigmoid Contrastive Loss
di: Bangachev, Kiril, et al.
Pubblicazione: (2025)

Theory, Analysis, and Best Practices for Sigmoid Self-Attention
di: Ramapuram, Jason, et al.
Pubblicazione: (2024)

Explainable and Interpretable Forecasts on Non-Smooth Multivariate Time Series for Responsible Gameplay
di: Jagirdar, Hussain, et al.
Pubblicazione: (2025)

Empirical evaluation of Time Series Foundation Models for Day-ahead and Imbalance Electricity Price Forecasting in Belgium
di: Bui, Chi, et al.
Pubblicazione: (2026)

Incorporating Retrieval-based Causal Learning with Information Bottlenecks for Interpretable Graph Neural Networks
di: Rao, Jiahua, et al.
Pubblicazione: (2024)