Search

Upcycling pre-trained dense language models into sparse mixture-of-experts (MoE) models is an efficient approach to increase the model capacity of already trained models. However, optimal techniques for upcycling at scale remain unclear. In this work, we conduct an extensive study of upcycling methods and hyperparameters for billion-parameter scale language models. We propose a novel "virtual group" initialization scheme and weight scaling approach to enable upcycling into fine-grained MoE architectures. Through ablations, we find that upcycling outperforms continued dense model training. In addition, we show that softmax-then-topK expert routing improves over topK-then-softmax approach and higher granularity MoEs can help improve accuracy. Finally, we upcycled Nemotron-4 15B on 1T tokens and compared it to a continuously trained version of the same model on the same 1T tokens: the continuous trained model achieved 65.3% MMLU, whereas the upcycled model achieved 67.6%. Our results offer insights and best practices to effectively leverage upcycling for building MoE language models.

Autonomous applications are typically developed over Robot Operating System 2.0 (ROS2) even in time-critical systems like automotive. Recent years have seen increased interest in developing model-based timing analysis and schedule optimization approaches for ROS2-based applications. To complement these approaches, we propose a tracing and measurement framework to obtain timing models of ROS2-based applications. It offers a tracer based on extended Berkeley Packet Filter (eBPF) that probes different functions in ROS2 middleware and reads their arguments or return values to reason about the data flow in applications. It combines event traces from ROS2 and the operating system to generate a directed acyclic graph showing ROS2 callbacks, precedence relations between them, and their timing attributes. While being compatible with existing analyses, we also show how to model (i)~message synchronization, e.g., in sensor fusion, and (ii)~service requests from multiple clients, e.g., in motion planning. Considering that, in real-world scenarios, the application code might be confidential and formal models are unavailable, our framework still enables the application of existing analysis and optimization techniques.

This paper proposes DisCo, an automatic deep learning compilation module for data-parallel distributed training. Unlike most deep learning compilers that focus on training or inference on a single device, DisCo optimizes a DNN model for distributed training over multiple GPU machines. Existing single-device compilation strategies do not work well in distributed training, due mainly to communication inefficiency that they incur. DisCo generates optimized, joint computation operator and communication tensor fusion strategies to enable highly efficient distributed training. A GNN-based simulator is built to effectively estimate per-iteration training time achieved by operator/tensor fusion candidates. A backtracking search algorithm is driven by the simulator, navigating efficiently in the large strategy space to identify good operator/tensor fusion strategies that minimize distributed training time. We compare DisCo with existing DL fusion schemes and show that it achieves good training speed-up close to the ideal, full computation-communication overlap case.

To train modern large DNN models, pipeline parallelism has recently emerged, which distributes the model across GPUs and enables different devices to process different microbatches in pipeline. Earlier pipeline designs allow multiple versions of model parameters to co-exist (similar to asynchronous training), and cannot ensure the same model convergence and accuracy performance as without pipelining. Synchronous pipelining has recently been proposed which ensures model performance by enforcing a synchronization barrier between training iterations. Nonetheless, the synchronization barrier requires waiting for gradient aggregation from all microbatches and thus delays the training progress. Optimized pipeline planning is needed to minimize such wait and hence the training time, which has not been well studied in the literature. This paper designs efficient, near-optimal algorithms for expediting synchronous pipeline-parallel training of modern large DNNs over arbitrary inter-GPU connectivity. Our algorithm framework comprises two components: a pipeline partition and device mapping algorithm, and a pipeline scheduler that decides processing order of microbatches over the partitions, which together minimize the per-iteration training time. We conduct thorough theoretical analysis, extensive testbed experiments and trace-driven simulation, and demonstrate our scheme can accelerate training up to 157% compared with state-of-the-art designs., Comment: INFOCOM 2022

Low-quality face image restoration is a popular research direction in today's computer vision field. It can be used as a pre-work for tasks such as face detection and face recognition. At present, there is a lot of work to solve the problem of low-quality faces under various environmental conditions. This paper mainly focuses on the restoration of motion-blurred faces. In increasingly abundant mobile scenes, the fast recovery of motion-blurred faces can bring highly effective speed improvements in tasks such as face matching. In order to achieve this goal, a deblurring method for motion-blurred facial image signals based on generative adversarial networks(GANs) is proposed. It uses an end-to-end method to train a sharp image generator, i.e., a processor for motion-blurred facial images. This paper introduce the processing progress of motion-blurred images, the development and changes of GANs and some basic concepts. After that, it give the details of network structure and training optimization design of the image processor. Then we conducted a motion blur image generation experiment on some general facial data set, and used the pairs of blurred and sharp face image data to perform the training and testing experiments of the processor GAN, and gave some visual displays. Finally, MTCNN is used to detect the faces of the image generated by the deblurring processor, and compare it with the result of the blurred image. From the results, the processing effect of the deblurring processor on the motion-blurred picture has a significant improvement both in terms of intuition and evaluation indicators of face detection., Comment: 10 pages, 4 figures, appear in International Conference on Digital Signal Processing (ICDSP) 2021

Convolutional neural networks (CNNs) have been used in many machine learning fields. In practical applications, the computational cost of convolutional neural networks is often high with the deepening of the network and the growth of data volume, mostly due to a large amount of multiplication operations of floating-point numbers in convolution operations. To reduce the amount of multiplications, we propose a new type of CNNs called Tropical Convolutional Neural Networks (TCNNs) which are built on tropical convolutions in which the multiplications and additions in conventional convolutional layers are replaced by additions and min/max operations respectively. In addition, since tropical convolution operators are essentially nonlinear operators, we expect TCNNs to have higher nonlinear fitting ability than conventional CNNs. In the experiments, we test and analyze several different architectures of TCNNs for image classification tasks in comparison with similar-sized conventional CNNs. The results show that TCNN can achieve higher expressive power than ordinary convolutional layers on the MNIST and CIFAR10 image data set. In different noise environments, there are wins and losses in the robustness of TCNN and ordinary CNNs., Comment: 10 pages, 4 figures, appear in International Conference on Compute and Data Analysis (ICCDA) 2021

We present a new model of neural networks called Min-Max-Plus Neural Networks (MMP-NNs) based on operations in tropical arithmetic. In general, an MMP-NN is composed of three types of alternately stacked layers, namely linear layers, min-plus layers and max-plus layers. Specifically, the latter two types of layers constitute the nonlinear part of the network which is trainable and more sophisticated compared to the nonlinear part of conventional neural networks. In addition, we show that with higher capability of nonlinearity expression, MMP-NNs are universal approximators of continuous functions, even when the number of multiplication operations is tremendously reduced (possibly to none in certain extreme cases). Furthermore, we formulate the backpropagation algorithm in the training process of MMP-NNs and introduce an algorithm of normalization to improve the rate of convergence in training., Comment: 21 pages, 5 figures

The last decade has witnessed growth in the computational requirements for training deep neural networks. Current approaches (e.g., data/model parallelism, pipeline parallelism) parallelize training tasks onto multiple devices. However, these approaches always rely on specific deep learning frameworks and requires elaborate manual design, which make it difficult to maintain and share between different type of models. In this paper, we propose Auto-MAP, a framework for exploring distributed execution plans for DNN workloads, which can automatically discovering fast parallelization strategies through reinforcement learning on IR level of deep learning models. Efficient exploration remains a major challenge for reinforcement learning. We leverage DQN with task-specific pruning strategies to help efficiently explore the search space including optimized strategies. Our evaluation shows that Auto-MAP can find the optimal solution in two hours, while achieving better throughput on several NLP and convolution models.

It is a challenging task to train large DNN models on sophisticated GPU platforms with diversified interconnect capabilities. Recently, pipelined training has been proposed as an effective approach for improving device utilization. However, there are still several tricky issues to address: improving computing efficiency while ensuring convergence, and reducing memory usage without incurring additional computing costs. We propose DAPPLE, a synchronous training framework which combines data parallelism and pipeline parallelism for large DNN models. It features a novel parallelization strategy planner to solve the partition and placement problems, and explores the optimal hybrid strategy of data and pipeline parallelism. We also propose a new runtime scheduling algorithm to reduce device memory usage, which is orthogonal to re-computation approach and does not come at the expense of training throughput. Experiments show that DAPPLE planner consistently outperforms strategies generated by PipeDream's planner by up to 3.23x under synchronous training scenarios, and DAPPLE runtime outperforms GPipe by 1.6x speedup of training throughput and reduces the memory consumption of 12% at the same time.