MLPerf benchmarks ML training to drive performance improvements.
problem Unique challenges in ML training benchmarks.
method Developed MLPerf to overcome ML training's specific challenges.
result Quantitatively evaluated MLPerf's effectiveness.
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
169,291 papers · 148 categories
Trend · papers per month
48 results for “Training Benchmark”
This work introduces a new benchmark to compare neural network training algorithms.
problem Lack of reliable benchmarks to compare training algorithms effectively.
method Developed a new benchmark called AlgoPerf: Training Algorithms benchmark.
result Demonstrated the feasibility of the benchmark and set a provisional state-of-the-art.
MLPerf benchmark suite evaluates diverse ML applications, highlighting system bottlenecks.
problem Understanding and optimizing ML applications across different models and systems.
method Analysis of MLPerf benchmark suite characteristics and comparison with previous benchmarks.
result Optimal distributed deep learning training requires dedicated low latency interconnects.
This paper benchmarks algorithms for training fair DNNs, addressing real-world fairness constraints.
problem Training deep neural networks with fairness constraints.
method Benchmarking stochastic approximation algorithms for fairness-constrained DNN training.
result Demonstrates the use of a new benchmark for comparing fairness-improving algorithms.
This work benchmarks and theorizes robust NAS under adversarial training.
problem Lack of benchmark evaluations and theoretical guarantees for robust NAS architectures under adversarial training.
method Released a comprehensive data set and established a generalization theory using the neural tangent kernel.
result Established a generalization theory for robust NAS architectures under adversarial training.
New benchmark protocol evaluates neural network optimizers for efficiency and data shift sensitivity.
problem Benchmarking neural network optimizers with hyperparameter complexity and data shift sensitivity.
method Proposed a new evaluation protocol combining end-to-end and data-addition training efficiency, using bandit hyperparameter tuning and human study validation.
result No clear winner across all tasks, highlighting the complexity of optimizer performance.
Procgen Benchmark uses procedurally generated games to test reinforcement learning.
problem Lack of diverse and high-quality training environments for reinforcement learning.
method Developed 16 procedurally generated game-like environments and used them to benchmark reinforcement learning.
result Procedurally generated environments are essential for training and evaluating reinforcement learning agents.
Sloth predicts LLM performance using latent skills across families.
problem Variations in benchmark performance due to differences in training configurations and data processing across model families.
method Sloth uses publicly available benchmark data and assumes LLM performance is driven by latent skills influenced by model size and training tokens. It exploits correlations across benchmarks to provide accurate predictions.
result Sloth predicts LLM performance accurately and offers insights into scaling behaviors for complex tasks.
BREEDS benchmarks assess model robustness to subpopulation shifts.
problem Measuring model robustness to novel subpopulation shifts.
method Controlled synthesis of realistic distribution shifts using class structure.
result Validated model sensitivity and effectiveness of robustness interventions.
New benchmarks measure image generation models' ability to generalize beyond training data.
problem Trivially memorizing training data yields better scores than state-of-the-art models on current benchmarks.
method Developed neural network divergences (NNDs) as evaluation metrics requiring large samples.
result Implemented and validated a black-box metric that measures diversity, sample quality, and generalization.
Proposes a new benchmark for deep neural network training.
problem Limited focus on DNN training efficiency.
method Develops TBD benchmark covering various applications and frameworks.
result Highlights inefficiencies in DNN training across different models and hardware.
New benchmarks provide full training data for NAS research.
problem Limited training data on popular benchmarks restricts multi-fidelity techniques.
method SVD and noise modeling to create surrogate benchmarks with full training info.
result Learning curve extrapolation framework improves single-fidelity algorithms.
MOSES benchmarks molecular generation models using a standardized dataset and metrics.
problem Unclear comparison and ranking of molecular generation models.
method Developed MOSES platform with training and testing datasets, metrics.
result Suggested MOSES results as reference for advancements in generative chemistry.
AVE measures redundancy in ligand-based benchmarks, revealing overfitting.
problem Overfitting in ligand-based classification benchmarks.
method AVE (Training-Validation Redundancy Measure) for ligand-based classification problems.
result Performance of ligand-based methods correlates with AVE bias, not generalization.
Paper proposes a new speech representation benchmark and model.
problem Lack of benchmarks for comparing speech representations.
method Unsupervised triplet-loss objective for training a universal non-semantic speech representation.
result Proposed representation outperforms other models on benchmark and transfer learning tasks.
Adversarial training effectiveness varies widely due to inconsistent training settings.
problem Variability in adversarial training effectiveness due to inconsistent training settings.
method Comprehensive evaluation of 10+ adversarial training methods and their hyperparameters.
result Basic training settings like weight decay can significantly impact adversarial robustness.
Fashion-MNIST replaces MNIST for machine learning benchmarks.
problem No new problem introduced.
method No new method introduced.
result Fashion-MNIST serves as a direct replacement for MNIST.
BloombergGPT is a large language model trained on financial data, outperforming existing models on financial tasks.
problem Lack of specialized large language models for finance.
method Trained on a 363 billion token dataset augmented with 345 billion tokens from general datasets, using a 50 billion parameter model.
result BloombergGPT outperforms existing models on financial tasks without sacrificing performance on general LLM benchmarks.
Reservoir Memory Machines solve benchmark tasks faster than Neural Turing Machines.
problem Training Neural Turing Machines is hard and limits their applicability.
method Proposes Reservoir Memory Machines, combining neural network flexibility with Turing machine capabilities, but with faster training via alignment and linear regression.
result Reservoir Memory Machines solve benchmark tasks as well as Neural Turing Machines but are much faster to train.
This paper predicts weekly stock market movements using machine learning and introduces a new benchmark.
problem Predicting stock market movements using daily data and various ML models.
method Focuses on weekly movements, introduces random traders as a benchmark, uses additional features, and adjusts training datasets.
result Trained models, especially MLP, show good performance across different trends.
New benchmark for non-rigid 3D human shape retrieval.
problem Distinguishing between body shapes of 3D human models.
method Extended benchmark with 145 new models and FAUST dataset.
result Improved comparison of 25 shape retrieval methods.
PINNs struggle with increasingly complex ODEs, especially when parameters control their complexity.
problem Evaluating physics-informed neural networks on complex coupled ODEs.
method Tuned benchmarks of partial differential equations and harmonic oscillators; varying network architecture and training method.
result PINNs fail to solve complex ODEs, revealing issues like insufficient capacity, poor conditioning, and high local curvature.
Agent57 beats human average on all Atari games.
problem Atari game benchmark's poor performance in challenging games.
method Trained neural network with adaptive policy selection and novel architecture.
result First deep RL agent to outperform human average on all 57 Atari games.
Benchmark proposes to assess molecule docking efficiency.
problem Lack of realistic benchmarks for measuring progress in drug design.
method Proposes a docking-based benchmark using SMINA software.
result Graph-based generative models fail to generate high-scoring molecules.
New HPO frameworks needed for continual learning.
problem No standard HPO for continual learning.
method Comparative study of HPO frameworks.
result No HPO framework consistently outperforms others.
Paper establishes a comprehensive benchmark for ECG time-series analysis.
problem Incomplete understanding of ECG signal properties and limitations in evaluation metrics.
method Categorization of downstream applications, identification of limitations, introduction of a novel metric, benchmarking of time-series models.
result Validation of the effectiveness of the proposed metric and model architecture.
This paper benchmarks batch RL algorithms on Atari, finding DQN and partially-trained policies perform best.
problem Deep RL algorithms fail in batch setting.
method Benchmarked batch RL algorithms on Atari using a single partially-trained policy.
result Many batch RL algorithms underperform DQN and partially-trained policies.
DAWNbench evaluates deep learning training time vs. accuracy, revealing hardware underutilization.
problem Lack of standard evaluation metrics for deep learning performance.
method Introduced DAWNBench, a benchmark focusing on training time to achieve near-state-of-the-art accuracy.
result Training time to accuracy (TTA) is a reliable metric for comparing deep learning optimizations.
A benchmark for NLP models trained on text datasets.
problem Limited access to high-performance clusters for NAS experiments.
method Created a search space for recurrent neural networks on text datasets and trained 14k architectures.
result Demonstrated the potential of precomputed NAS results for NLP.
Public dataset for benchmarking deep learning CT reconstruction methods.
problem Lack of a fair benchmark for comparing deep learning CT reconstruction methods.
method Processed and simulated over 40,000 CT scan slices from the LIDC/IDRI Database.
result First baseline results provided for comparison.
Enhances graph classification models on small datasets.
problem Over-fitting and undergeneralization on small-scale benchmark datasets.
method Data augmentation via graph structure transformation and model evolution framework.
result Average improvement of 3 - 13% accuracy on graph classification tasks.
WILDS 2.0 expands benchmark datasets for unsupervised adaptation.
problem Leveraging unlabeled data for distribution shifts in real-world applications.
method Curated unlabeled data across various applications, tasks, and modalities.
result State-of-the-art methods perform poorly on WILDS datasets.
Fidel-TS creates a new benchmark for time series forecasting models.
problem Lack of high-quality benchmarks for time series forecasting models.
method Formalized high-fidelity benchmark principles, including data sourcing integrity, leak-free design, and structural clarity. Created Fidel-TS, a new large-scale benchmark.
result Demonstrated the limitations of prior benchmarks and potential discrepancies in model evaluation.
AI benchmarks evaluate football team performance using generative models.
problem Evaluating human performance in complex interactive tasks is error-prone and unreliable.
method Trained Conditional VRNN Model on player and ball tracking data to imitate and predict team interactions.
result Trained model as a useful benchmark for evaluating team performance in football.
New benchmarks for offline RL from diverse datasets.
problem Measuring progress in offline RL due to lack of suitable benchmarks.
method Developed benchmarks tailored for offline RL, focusing on diverse dataset properties.
result Revealed deficiencies in existing offline RL algorithms.
Study benchmarks machine learning models for fake news detection.
problem Dataset bias and performance of fake news detection models.
method Benchmarked different machine learning models on three datasets.
result BERT and similar pre-trained models perform best for fake news detection.
L2O uses machine learning to design optimization methods.
problem Designing efficient optimization methods for specific problem distributions.
method Data-driven approach to automate optimization method design.
result L2O methods are practical for specific problem distributions but fail on out-of-distribution problems.
The paper aims to define a benchmark for deep learning recommendation models.
problem Insufficient benchmarking for deep learning recommendation models.
method Synthesizes modeling strategies, defines desirable characteristics, and summarizes advice from the MLPerf Recommendation Advisory Board.
result Defines an industry-relevant benchmark for deep learning recommendation models.
Efficiently trains GPs with billions of inducing inputs using Tensor Train decomposition.
problem Training GPs with large numbers of inducing inputs.
method Tensor Train decomposition for variational parameters in stochastic variational inference.
result Achieves state-of-the-art results on several benchmarks.
Study benchmarks label noise detection methods, identifying best practices.
problem Label noise in real-world datasets affects model performance and evaluation reliability.
method Decomposed detection methods into label agreement, aggregation, and information gathering components; introduced a unified benchmark task and novel metric.
result In-sample probability aggregation with logit margin label agreement function achieves best results across scenarios.
New benchmarks improve model performance by accounting for isomorphism classes in multi-relational datasets.
problem Synthetic multi-relational datasets lack isomorphism class awareness, leading to overestimation of model performance.
method Proposed isomorphism-aware synthetic benchmarks and a prioritisation scheme to improve model performance and stability.
result Isomorphism classes can be utilised to improve model performance, stability during training, and reduce training time.
New framework selects high-quality pretraining data without training LLMs.
problem Slow progress in understanding pretraining data due to costly experiments.
method Statistical framework based on perplexity-benchmark correlations.
result Approach outperforms existing methods on multiple benchmarks.
Efficient echo state network with explicit memory performs well on benchmark tasks.
problem Training differentiable neural computers is difficult and time-consuming.
method Echo state network with an explicit memory.
result Echo state network can recognize all regular languages, including those contractive networks cannot.
We argue for the principle of unchanged optimality in RL benchmarks and discuss its implications.
problem Generalization in reinforcement learning benchmarks.
method Discussion of conceptual properties and subtle choices in state representation and model architecture.
result The principle of unchanged optimality is important for RL benchmarks and can be broken or satisfied by model architecture choices.
Facebook's ResNeXt WSL models show exceptional robustness against image corruptions and adversarial attacks.
problem Image recognition model robustness against corruptions and adversarial attacks.
method Training with 1B images from Instagram and fine-tuning on ImageNet.
result ResNeXt WSL models achieve state-of-the-art results on ImageNet-C, ImageNet-P, and ImageNet-A.
Proposes a method to evaluate meta-learning performance based on task similarity.
problem Meta-learning performance evaluation ignores task similarity, leading to biased results.
method Generative approach using Latent Dirichlet Allocation to analyze task similarity.
result The proposed method provides an insightful evaluation of meta-learning algorithms, matching common intuition.
COPML framework securely trains models across multiple data owners without revealing individual data.
problem Privacy-preserving collaborative machine learning with multiple data owners.
method Securely encodes data, distributes computation, performs distributed training.
result Achieves up to 16x speedup in training time while maintaining strong privacy.
CogDL simplifies graph deep learning experiments and benchmarks.
problem Challenges in training and evaluating graph neural networks.
method Unified design for training and evaluation, mixed precision training, efficient sparse operators, ease of use.
result CogDL is the most competitive graph library for efficiency and ease of use.