Cover Image

Understanding the Training Speedup from Sampling with Approximate Losses

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Das, Rudrajit, Chen, Xi, Ieong, Bertram, Bansal, Parikshit, Sanghavi, Sujay
Format:	Preprint
Published:	2024
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Enabling Approximate Joint Sampling in Diffusion LMs
by: Bansal, Parikshit, et al.
Published: (2025)

Context-Free Synthetic Data Mitigates Forgetting
by: Bansal, Parikshit, et al.
Published: (2025)

Understanding Self-Supervised Learning via Gaussian Mixture Models
by: Bansal, Parikshit, et al.
Published: (2024)

Upweighting Easy Samples in Fine-Tuning Mitigates Forgetting
by: Sanyal, Sunny, et al.
Published: (2025)

Towards Quantifying the Preconditioning Effect of Adam
by: Das, Rudrajit, et al.
Published: (2024)

Test-Time Speculation
by: Kumar, Avinash, et al.
Published: (2026)

Retraining with Predicted Hard Labels Provably Increases Model Accuracy
by: Das, Rudrajit, et al.
Published: (2024)

HiSpec: Hierarchical Speculative Decoding for LLMs
by: Kumar, Avinash, et al.
Published: (2025)

Learning Mixtures of Experts with EM: A Mirror Descent Perspective
by: Fruytier, Quentin, et al.
Published: (2024)

Geometric Median (GM) Matching for Robust Data Pruning
by: Acharya, Anish, et al.
Published: (2024)

Self-Boost via Optimal Retraining: An Analysis via Approximate Message Passing
by: Javanmard, Adel, et al.
Published: (2025)

Blocking Bandits
by: Basu, Soumya, et al.
Published: (2019)

Geometric Median Matching for Robust k-Subset Selection from Noisy Data
by: Acharya, Anish, et al.
Published: (2025)

Asymptotically-Optimal Gaussian Bandits with Side Observations
by: Atsidakou, Alexia, et al.
Published: (2025)

Finite-Time Logarithmic Bayes Regret Upper Bounds
by: Atsidakou, Alexia, et al.
Published: (2023)

Small LLMs with Expert Blocks Are Good Enough for Hyperparamter Tuning
by: Naphade, Om, et al.
Published: (2025)

Understanding Contrastive Representation Learning from Positive Unlabeled (PU) Data
by: Acharya, Anish, et al.
Published: (2024)

Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization
by: Jiang, Ruichen, et al.
Published: (2024)

Less is More: Convergence Benefits of Fewer Data Weight Updates over Longer Horizon
by: Das, Rudrajit, et al.
Published: (2026)

Time Weaver: A Conditional Time Series Generation Model
by: Narasimhan, Sai Shankar, et al.
Published: (2024)

When Attention Collapses: How Degenerate Layers in LLMs Enable Smaller, Stronger Models
by: Sanyal, Sunny, et al.
Published: (2024)

In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness
by: Collins, Liam, et al.
Published: (2024)

Entropy Aware Reward Guidance for Diffusion Language Model Alignment
by: Tejaswi, Atula, et al.
Published: (2026)

Understanding Uncertainty Sampling via Equivalent Loss
by: Liu, Shang, et al.
Published: (2023)

Beyond Speedup -- Utilizing KV Cache for Sampling and Reasoning
by: Xing, Zeyu, et al.
Published: (2026)

Pretrained deep models outperform GBDTs in Learning-To-Rank under label scarcity
by: Hou, Charlie, et al.
Published: (2023)

Sculpting Latent Spaces With MMD: Disentanglement With Programmable Priors
by: Fruytier, Quentin, et al.
Published: (2025)

Quantum Speedup for Spectral Approximation of Kronecker Products
by: Gao, Yeqi, et al.
Published: (2024)

Positive Unlabeled Contrastive Learning
by: Acharya, Anish, et al.
Published: (2022)

From Edges to Depth: Probing the Spatial Hierarchy in Vision Transformers
by: Sanghavi, Jainum
Published: (2026)

Efficiently Training Neural Networks for Imperfect Information Games by Sampling Information Sets
by: Bertram, Timo, et al.
Published: (2024)

Power Flow Approximations for Multiphase Distribution Networks using Gaussian Processes
by: Glover, Daniel, et al.
Published: (2025)

EPiC: Towards Lossless Speedup for Reasoning Training through Edge-Preserving CoT Condensation
by: Jia, Jinghan, et al.
Published: (2025)

Sampling and Loss Weights in Multi-Domain Training
by: Salmani, Mahdi, et al.
Published: (2025)

InfoPO: On Mutual Information Maximization for Large Language Model Alignment
by: Xiao, Teng, et al.
Published: (2025)

MURPHY: Feedback-Aware GRPO with Retrospective Credit Assignment for Multi-Turn Code Generation
by: Ekbote, Chanakya, et al.
Published: (2025)

Unified Graph Networks (UGN): A Deep Neural Framework for Solving Graph Problems
by: Dawn, Rudrajit, et al.
Published: (2025)

Approximate Equivariance in Reinforcement Learning
by: Park, Jung Yeon, et al.
Published: (2024)

Sample Complexity of Causal Identification with Temporal Heterogeneity
by: Rathod, Ameya, et al.
Published: (2026)

Towards Understanding the Influence of Training Samples on Explanations
by: Artelt, André, et al.
Published: (2024)