Scalable Hyperparameter-Divergent Ensemble Training with Automatic Learning Rate Exploration for Large Models

• URL: http://arxiv.org/abs/2604.24708v1
• Date: Mon, 27 Apr 2026 17:17:28 GMT
• Title: Scalable Hyperparameter-Divergent Ensemble Training with Automatic Learning Rate Exploration for Large Models
• Abstract summary: Training large neural networks with data-parallel gradient descent allocates N GPU replicas to compute effectively identical updates.<n>We propose a method that repurposes these replicas for simultaneous learning rate configurations overhead.<n>We further propose an automatic learning rate (autoLR) controller that treats the relative training loss across replicas as a performance signal.
• Score: 8.687019807776881
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Training large neural networks with data-parallel stochastic gradient descent allocates N GPU replicas to compute effectively identical updates -- a practice that leaves the rich space of learning rate configurations entirely unexplored during training. We propose Hyperparameter-Divergent Ensemble Training (HDET), a method that repurposes these replicas for simultaneous learning rate exploration at negligible communication overhead. HDET operates in alternating phases: a fan-out stage in which replicas train independently under a structured, symmetric spread of learning rates, and a converge stage in which parameters are averaged across all replicas via AllReduce every T steps. Building on this ensemble substrate, we further propose an automatic learning rate (auto-LR) controller that treats the relative training loss across replicas as a performance signal, updating the shared base schedule toward higher-performing configurations via a momentum-based gradient-free meta-update. The combined method produces a self-adapting learning rate schedule that improves both optimization quality and generalization without additional hyperparameter sweeps or training budget. Crucially, the framework generalizes beyond learning rate: any scalar hyperparameter that does not alter model architecture -- such as dropout rate, attention scale temperature, or weight-decay coefficient -- can be explored across replicas using the same fan-out/converge protocol, with inter-replica loss differences serving as zero-order hypergradients that guide the search direction. HDET is implemented as a drop-in replacement for PyTorch's OneCycleLR scheduler, requiring no changes to model architecture, optimizer, or data pipeline.

Related papers

• Neural Network Optimization Reimagined: Decoupled Techniques for Scratch and Fine-Tuning [49.751529745537546]
  We propose DualOpt, a novel approach that decouples optimization techniques specifically tailored for training from scratch.<n>For training from scratch, we introduce real-time layer-wise weight decay, designed to enhance both convergence and generalization.<n>We extend the layer-wise weight decay to dynamically adjust the rollback levels across layers, adapting to the varying demands of different downstream tasks.
  arXiv  Detail & Related papers  (2026-04-21T06:27:18Z)
• SIMPLER: Efficient Foundation Model Adaptation via Similarity-Guided Layer Pruning for Earth Observation [3.1313965020437546]
  We introduce SIMPLER, a pre-fine-tuning architecture selection method that reduces inference and deployment costs.<n>On Prithvi-EO-2, SIMPLER prunes up to 79% of parameters while retaining 94% of baseline performance, yielding a 2.1x training speedup and 2.6x inference speedup.
  arXiv  Detail & Related papers  (2026-03-20T11:38:32Z)
• DRL: Discriminative Representation Learning with Parallel Adapters for Class Incremental Learning [63.65467569295623]
  We propose the Discriminative Representation Learning (DRL) framework to specifically address these challenges.<n>To conduct incremental learning effectively and yet efficiently, the DRL's network is built upon a PTM.<n>Our DRL consistently outperforms other state-of-the-art methods throughout the entire CIL period.
  arXiv  Detail & Related papers  (2025-10-14T03:19:15Z)
• Sig2Model: A Boosting-Driven Model for Updatable Learned Indexes [6.133666849556217]
  Sig2Model is an efficient and adaptive learned index that minimizes retraining cost through three key techniques.<n>We show that Sig2Model reduces retraining cost by up to 20x, achieves up to 3x higher QPS, and uses up to 1000x less memory.
  arXiv  Detail & Related papers  (2025-09-25T06:07:13Z)
• Protocol Models: Scaling Decentralized Training with Communication-Efficient Model Parallelism [59.79227116582264]
  Scaling models has led to significant advancements in deep learning, but training these models in decentralized settings remains challenging.<n>We propose a novel compression algorithm that compresses both forward and backward passes, enabling up to 99% compression with no convergence degradation.
  arXiv  Detail & Related papers  (2025-06-02T02:19:22Z)
• ALoRE: Efficient Visual Adaptation via Aggregating Low Rank Experts [71.91042186338163]
  ALoRE is a novel PETL method that reuses the hypercomplex parameterized space constructed by Kronecker product to Aggregate Low Rank Experts.<n>Thanks to the artful design, ALoRE maintains negligible extra parameters and can be effortlessly merged into the frozen backbone.
  arXiv  Detail & Related papers  (2024-12-11T12:31:30Z)
• SHERL: Synthesizing High Accuracy and Efficient Memory for Resource-Limited Transfer Learning [63.93193829913252]
  We propose an innovative METL strategy called SHERL for resource-limited scenarios. In the early route, intermediate outputs are consolidated via an anti-redundancy operation. In the late route, utilizing minimal late pre-trained layers could alleviate the peak demand on memory overhead.
  arXiv  Detail & Related papers  (2024-07-10T10:22:35Z)
• Online Convolutional Re-parameterization [51.97831675242173]
  We present online convolutional re- parameterization (OREPA), a two-stage pipeline, aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution. Compared with the state-of-the-art re-param models, OREPA is able to save the training-time memory cost by about 70% and accelerate the training speed by around 2x. We also conduct experiments on object detection and semantic segmentation and show consistent improvements on the downstream tasks.
  arXiv  Detail & Related papers  (2022-04-02T09:50:19Z)
• Scalable One-Pass Optimisation of High-Dimensional Weight-Update Hyperparameters by Implicit Differentiation [0.0]
  We develop an approximate hypergradient-based hyper parameter optimiser. It requires only one training episode, with no restarts. We also provide a motivating argument for convergence to the true hypergradient.
  arXiv  Detail & Related papers  (2021-10-20T09:57:57Z)
• AgEBO-Tabular: Joint Neural Architecture and Hyperparameter Search with Autotuned Data-Parallel Training for Tabular Data [11.552769149674544]
  Development of high-performing predictive models for large data sets is a challenging task. Recent automated machine learning (AutoML) is emerging as a promising approach to automate predictive model development. We have developed AgEBO-Tabular, an approach to combine aging evolution (AgE) and a parallel NAS method that searches over neural architecture space.
  arXiv  Detail & Related papers  (2020-10-30T16:28:48Z)
• Robust Federated Learning Through Representation Matching and Adaptive Hyper-parameters [5.319361976450981]
  Federated learning is a distributed, privacy-aware learning scenario which trains a single model on data belonging to several clients. Current federated learning methods struggle in cases with heterogeneous client-side data distributions. We propose a novel representation matching scheme that reduces the divergence of local models.
  arXiv  Detail & Related papers  (2019-12-30T20:19:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.