NSML: Meet the MLaaS platform with a real-world case study

Abstract

The boom of deep learning induced many industries and academies to introduce machine learning based approaches into their concern, competitively. However, existing machine learning frameworks are limited to sufficiently fulfill the collaboration and management for both data and models. We proposed NSML, a machine learning as a service (MLaaS) platform, to meet these demands. NSML helps machine learning work be easily launched on a NSML cluster and provides a collaborative environment which can afford development at enterprise scale. Finally, NSML users can deploy their own commercial services with NSML cluster. In addition, NSML furnishes convenient visualization tools which assist the users in analyzing their work. To verify the usefulness and accessibility of NSML, we performed some experiments with common examples. Furthermore, we examined the collaborative advantages of NSML through three competitions with real-world use cases.

References in corpus (8)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Deep Residual Learning for Image Recognition
• An introduction to Docker for reproducible research, with examples from the R environment
• Random Erasing Data Augmentation
• The Microsoft 2017 Conversational Speech Recognition System
• QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
• ShakeDrop Regularization for Deep Residual Learning
• Variational Autoencoders for Collaborative Filtering

Cited by in corpus (49)

• CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
• DialogWAE: Multimodal Response Generation with Conditional Wasserstein Auto-Encoder
• StarGAN v2: Diverse Image Synthesis for Multiple Domains
• AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights
• What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis
• Rethinking Spatial Dimensions of Vision Transformers
• Reliable Fidelity and Diversity Metrics for Generative Models
• Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram
• VideoMix: Rethinking Data Augmentation for Video Classification
• Self-supervised Auxiliary Learning with Meta-paths for Heterogeneous Graphs
• Photorealistic Style Transfer via Wavelet Transforms
• An Empirical Evaluation on Robustness and Uncertainty of Regularization Methods
• EXTD: Extremely Tiny Face Detector via Iterative Filter Reuse
• Toward Interpretable Music Tagging with Self-Attention
• HyperTendril: Visual Analytics for User-Driven Hyperparameter Optimization of Deep Neural Networks
• Tripartite Heterogeneous Graph Propagation for Large-scale Social Recommendation
• Evaluating Weakly Supervised Object Localization Methods Right
• Efficient Dialogue State Tracking by Selectively Overwriting Memory
• Rainbow Memory: Continual Learning with a Memory of Diverse Samples
• Probability density distillation with generative adversarial networks for high-quality parallel waveform generation
• Rethinking Channel Dimensions for Efficient Model Design
• Orchestrating the Development Lifecycle of Machine Learning-Based IoT Applications: A Taxonomy and Survey
• A Comprehensive Overhaul of Feature Distillation
• Length-Adaptive Transformer: Train Once with Length Drop, Use Anytime with Search
• Multi-Manifold Learning for Large-scale Targeted Advertising System
• Reward Optimization for Neural Machine Translation with Learned Metrics
• Mixture Content Selection for Diverse Sequence Generation
• ClovaCall: Korean Goal-Oriented Dialog Speech Corpus for Automatic Speech Recognition of Contact Centers
• HPCClusterScape: Increasing Transparency and Efficiency of Shared High-Performance Computing Clusters for Large-scale AI Models
• Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
• Probabilistic Embeddings for Cross-Modal Retrieval
• Which Strategies Matter for Noisy Label Classification? Insight into Loss and Uncertainty
• Boosting Active Learning for Speech Recognition with Noisy Pseudo-labeled Samples
• SWAD: Domain Generalization by Seeking Flat Minima
• Self-supervised Auxiliary Learning for Graph Neural Networks via Meta-Learning
• FrostNet: Towards Quantization-Aware Network Architecture Search
• ST-BERT: Cross-modal Language Model Pre-training For End-to-end Spoken Language Understanding
• StyleAugment: Learning Texture De-biased Representations by Style Augmentation without Pre-defined Textures
• Scaling Law for Recommendation Models: Towards General-purpose User Representations
• Which Ads to Show? Advertisement Image Assessment with Auxiliary Information via Multi-step Modality Fusion
• Hop Sampling: A Simple Regularized Graph Learning for Non-Stationary Environments
• M2FN: Multi-step Modality Fusion for Advertisement Image Assessment
• Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editing
• Two-stage Textual Knowledge Distillation for End-to-End Spoken Language Understanding
• Graphs, Entities, and Step Mixture
• Subword Language Model for Query Auto-Completion
• Designing a Minimal Retrieve-and-Read System for Open-Domain Question Answering
• NeuralWOZ: Learning to Collect Task-Oriented Dialogue via Model-Based Simulation
• Large Product Key Memory for Pretrained Language Models