Copertina

There is a Singularity in the Loss Landscape

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autore principale:	Lowell, Mark
Natura:	Preprint
Pubblicazione:	2022
Soggetti:	Machine Learning Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Landscaper: Understanding Loss Landscapes Through Multi-Dimensional Topological Analysis
di: Chen, Jiaqing, et al.
Pubblicazione: (2026)

Evaluating Loss Landscapes from a Topology Perspective
di: Xie, Tiankai, et al.
Pubblicazione: (2024)

Visualizing Loss Functions as Topological Landscape Profiles
di: Geniesse, Caleb, et al.
Pubblicazione: (2024)

Model Merging on Loss Landscape: A Geometry Perspective
di: Lu, Juanwu, et al.
Pubblicazione: (2026)

Early-Warning Signals of Grokking via Loss-Landscape Geometry
di: Xu, Yongzhong
Pubblicazione: (2026)

Matrix Sensing with Kernel Optimal Loss: Robustness and Optimization Landscape
di: Song, Xinyuan, et al.
Pubblicazione: (2025)

Identifying Sparsely Active Circuits Through Local Loss Landscape Decomposition
di: Chrisman, Brianna, et al.
Pubblicazione: (2025)

A Tale of Two Symmetries: Exploring the Loss Landscape of Equivariant Models
di: Xie, YuQing, et al.
Pubblicazione: (2025)

Practical Bayesian Inference for Speech SNNs: Uncertainty and Loss-Landscape Smoothing
di: Abdennadher, Yesmine, et al.
Pubblicazione: (2026)

Singular Vectors of Attention Heads Align with Features
di: Franco, Gabriel, et al.
Pubblicazione: (2026)

Loss Landscape Degeneracy and Stagewise Development in Transformers
di: Hoogland, Jesse, et al.
Pubblicazione: (2024)

Flat Channels to Infinity in Neural Loss Landscapes
di: Martinelli, Flavio, et al.
Pubblicazione: (2025)

Adapting Critic Match Loss Landscape Visualization to Off-policy Reinforcement Learning
di: Liu, Jingyi, et al.
Pubblicazione: (2026)

Loss Barcode: A Topological Measure of Escapability in Loss Landscapes
di: Barannikov, Serguei, et al.
Pubblicazione: (2020)

Visualizing Critic Match Loss Landscapes for Interpretation of Online Reinforcement Learning Control Algorithms
di: Liu, Jingyi, et al.
Pubblicazione: (2026)

A Loss Landscape Visualization Framework for Interpreting Reinforcement Learning: An ADHDP Case Study
di: Liu, Jingyi, et al.
Pubblicazione: (2026)

The Optimiser Hidden in Plain Sight: Training with the Loss Landscape's Induced Metric
di: Harvey, Thomas R.
Pubblicazione: (2025)

Probing Implicit Bias in Semi-gradient Q-learning: Visualizing the Effective Loss Landscapes via the Fokker--Planck Equation
di: Yin, Shuyu, et al.
Pubblicazione: (2024)

M-HOF-Opt: Multi-Objective Hierarchical Output Feedback Optimization via Multiplier Induced Loss Landscape Scheduling
di: Sun, Xudong, et al.
Pubblicazione: (2024)

A Unified Noise-Curvature View of Loss of Trainability
di: Baveja, Gunbir Singh, et al.
Pubblicazione: (2025)

PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language Models
di: Meng, Fanxu, et al.
Pubblicazione: (2024)

Fractal Landscapes in Policy Optimization
di: Wang, Tao, et al.
Pubblicazione: (2023)

Dispelling the Curse of Singularities in Neural Network Optimizations
di: Cao, Hengjie, et al.
Pubblicazione: (2026)

Enhancing Graph Classification Robustness with Singular Pooling
di: Ennadir, Sofiane, et al.
Pubblicazione: (2025)

Problematizing AI Omnipresence in Landscape Architecture
di: Fernberg, Phillip, et al.
Pubblicazione: (2024)

ANO : Faster is Better in Noisy Landscape
di: Kegreisz, Adrien
Pubblicazione: (2025)

A Generalized Singular Value Theory for Neural Networks
di: Brown, Brian Charles, et al.
Pubblicazione: (2026)

A Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs
di: Kalra, Dayal Singh, et al.
Pubblicazione: (2026)

CP Loss: Channel-wise Perceptual Loss for Time Series Forecasting
di: Zha, Yaohua, et al.
Pubblicazione: (2026)

Singular Value Decomposition on Kronecker Adaptation for Large Language Model
di: Chong, Yee Hin, et al.
Pubblicazione: (2025)

Elliptic Loss Regularization
di: Hasan, Ali, et al.
Pubblicazione: (2025)

When Losses Align: Gradient-Based Composite Loss Weighting for Efficient Pretraining
di: Karpukhin, Ivan, et al.
Pubblicazione: (2026)

Robust Model-Based Optimization for Challenging Fitness Landscapes
di: Ghaffari, Saba, et al.
Pubblicazione: (2023)

Learning Advanced Self-Attention for Linear Transformers in the Singular Value Domain
di: Wi, Hyowon, et al.
Pubblicazione: (2025)

SPA-Cache: Singular Proxies for Adaptive Caching in Diffusion Language Models
di: Sun, Wenhao, et al.
Pubblicazione: (2026)

Investigating the Histogram Loss in Regression
di: Imani, Ehsan, et al.
Pubblicazione: (2024)

Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomy
di: Shaik, Thanveer, et al.
Pubblicazione: (2023)

Exploring the Landscape for Generative Sequence Models for Specialized Data Synthesis
di: Zbeeb, Mohammad, et al.
Pubblicazione: (2024)

GADPN: Graph Adaptive Denoising and Perturbation Networks via Singular Value Decomposition
di: Deng, Hao, et al.
Pubblicazione: (2026)

FANoise: Singular Value-Adaptive Noise Modulation for Robust Multimodal Representation Learning
di: Li, Jiaoyang, et al.
Pubblicazione: (2025)