相关论文: Oobleck: Resilient Distributed Training of Large M…

Can we preserve the accuracy of neural models while also providing faithful explanations of model decisions to training data? We propose a "wrapper box'' pipeline: training a neural model as usual and then using its learned feature…

The emergence of Deep Neural Networks (DNNs) in mission- and safety-critical applications brings their reliability to the front. High performance demands of DNNs require the use of specialized hardware accelerators. Systolic array…

Edge inference has become more widespread, as its diverse applications range from retail to wearable technology. Clusters of networked resource-constrained edge devices are becoming common, yet no system exists to split a DNN across these…

The customizable nature of deep learning models have allowed them to be successful predictors in various disciplines. These models are often trained with respect to thousands or millions of instances for complicated problems, but the…

Turbulent concentric coaxial pipe flows are numerically investigated as canonical problem addressing spanwise curvature effects on heat and momentum transfer that are encountered in various engineering applications. It is demonstrated that…

The effectiveness of the machine learning methods for real-world tasks depends on the proper structure of the modeling pipeline. The proposed approach is aimed to automate the design of composite machine learning pipelines, which is…

With the increasing scale of models, the need for efficient distributed training has become increasingly urgent. Recently, many synchronous pipeline parallelism approaches have been proposed to improve training throughput. However, these…

This work proposes the first strategy to make distributed training of neural networks resilient to computing errors, a problem that has remained unsolved despite being first posed in 1956 by von Neumann. He also speculated that the…

Binarization is an attractive strategy for implementing lightweight Deep Convolutional Neural Networks (CNNs). Despite the unquestionable savings offered, memory footprint above all, it may induce an excessive accuracy loss that prevents a…

Pipeline parallelism is a key technique for scaling large-model training, but modern workloads exhibit runtime variability in computation and communication. Existing pipeline systems typically consume static, profiled, or adaptively…

Conventional DNN training paradigms typically rely on one training set and one validation set, obtained by partitioning an annotated dataset used for training, namely gross training set, in a certain way. The training set is used for…

The size of Transformer models is growing at an unprecedented pace. It has only taken less than one year to reach trillion-level parameters after the release of GPT-3 (175B). Training such models requires both substantial engineering…

As Deep Neural Networks (DNNs) are rapidly being adopted within large software systems, software developers are increasingly required to design, train, and deploy such models into the systems they develop. Consequently, testing and…

Large machine learning models are revolutionary technologies of artificial intelligence whose bottlenecks include huge computational expenses, power, and time used both in the pre-training and fine-tuning process. In this work, we show that…

Deep neural networks achieve superior performance for learning from independent and identically distributed (i.i.d.) data. However, their performance deteriorates significantly when handling out-of-distribution (OoD) data, where the…

Although pretrained Transformers such as BERT achieve high accuracy on in-distribution examples, do they generalize to new distributions? We systematically measure out-of-distribution (OOD) generalization for seven NLP datasets by…

Reinforcement learning (RL) has become the core post-training technique for large language models (LLMs). RL for LLMs involves two stages: generation and training. The LLM first generates samples online, which are then used to derive…

Efficient Federated learning (FL) is crucial for training deep networks over devices with limited compute resources and bounded networks. With the advent of big data, devices either generate or collect multimodal data to train either…

Supercomputing systems today often come in the form of large numbers of commodity systems linked together into a computing cluster. These systems, like any distributed system, can have large numbers of independent hardware components…

Deep learning for distribution grid optimization can be advocated as a promising solution for near-optimal yet timely inverter dispatch. The principle is to train a deep neural network (DNN) to predict the solutions of an optimal power flow…