ForestColl: Efficient Collective Communications on Heterogeneous Network Fabrics

As modern DNN models grow ever larger, collective communications between the accelerators (allreduce, etc.) emerge as a significant performance bottleneck. Designing efficient communication schedules is challenging given today's highly diverse and heterogeneous network fabrics. In this paper, we present ForestColl, a tool that generates efficient schedules for any network topology. ForestColl constructs broadcast/aggregation spanning trees as the communication schedule, achieving theoretically minimum network congestion. Its schedule generation runs in strongly polynomial time and is highly scalable. ForestColl supports any network fabrics, including both switching fabrics and direct connections, as well as any network graph structure. We evaluated ForestColl on multi-cluster AMD MI250 and NVIDIA A100 platforms. ForestColl's schedules achieved up to 52\% higher performance compared to the vendors' own optimized communication libraries, RCCL and NCCL. ForestColl also outperforms other state-of-the-art schedule generation techniques with both up to 61\% more efficient generated schedules and orders of magnitude faster schedule generation speed.

FP8-LM: Training FP8 Large Language Models

In this paper, we explore FP8 low-bit data formats for efficient training of large language models (LLMs). Our key insight is that most variables, such as gradients and optimizer states, in LLM training can employ low-precision data formats without compromising model accuracy and requiring no changes to hyper-parameters. Specifically, we propose a new FP8 automatic mixed-precision framework for training LLMs. This framework offers three levels of FP8 utilization to streamline mixed-precision and distributed parallel training for LLMs. It gradually incorporates 8-bit gradients, optimizer states, and distributed learning in an incremental manner. Experiment results show that, during the training of GPT-175B model on H100 GPU platform, our FP8 mixed-precision training framework not only achieved a remarkable 42% reduction in real memory usage but also ran 64% faster than the widely adopted BF16 framework (i.e., Megatron-LM), surpassing the speed of Nvidia Transformer Engine by 17%. This largely reduces the training costs for large foundation models. Furthermore, our FP8 mixed-precision training methodology is generic. It can be seamlessly applied to other tasks such as LLM instruction tuning and reinforcement learning with human feedback, offering savings in fine-tuning expenses. Our FP8 low-precision training framework is open-sourced at {https://github.com/Azure/MS-AMP}{aka.ms/MS.AMP}.

Copyright Protection and Accountability of Generative AI:Attack, Watermarking and Attribution

Generative AI (e.g., Generative Adversarial Networks - GANs) has become increasingly popular in recent years. However, Generative AI introduces significant concerns regarding the protection of Intellectual Property Rights (IPR) (resp. model accountability) pertaining to images (resp. toxic images) and models (resp. poisoned models) generated. In this paper, we propose an evaluation framework to provide a comprehensive overview of the current state of the copyright protection measures for GANs, evaluate their performance across a diverse range of GAN architectures, and identify the factors that affect their performance and future research directions. Our findings indicate that the current IPR protection methods for input images, model watermarking, and attribution networks are largely satisfactory for a wide range of GANs. We highlight that further attention must be directed towards protecting training sets, as the current approaches fail to provide robust IPR protection and provenance tracing on training sets.

Tutel: Adaptive Mixture-of-Experts at Scale

In recent years, Mixture-of-Experts (MoE) has emerged as a promising technique for deep learning that can scale the model capacity to trillion-plus parameters while reducing the computing cost via sparse computation. While MoE opens a new frontier of exceedingly large models, its implementation over thousands of GPUs has been limited due to mismatch between the dynamic nature of MoE and static parallelism/pipelining of the system. We present Tutel, a highly scalable stack design and implementation for MoE with dynamically adaptive parallelism and pipelining. Tutel delivers adaptive parallelism switching and adaptive pipelining at runtime, which achieves up to 1.74x and 2.00x single MoE layer speedup, respectively. We also propose a novel two-dimensional hierarchical algorithm for MoE communication speedup that outperforms the previous state-of-the-art up to 20.7x over 2,048 GPUs. Aggregating all techniques, Tutel finally delivers 4.96x and 5.75x speedup of a single MoE layer on 16 GPUs and 2,048 GPUs, respectively, over Fairseq: Meta's Facebook AI Research Sequence-to-Sequence Toolkit (Tutel is now partially adopted by Fairseq). Tutel source code is available in public: https://github.com/microsoft/tutel . Our evaluation shows that Tutel efficiently and effectively runs a real-world MoE-based model named SwinV2-MoE, built upon Swin Transformer V2, a state-of-the-art computer vision architecture. On efficiency, Tutel accelerates SwinV2-MoE, achieving up to 1.55x and 2.11x speedup in training and inference over Fairseq, respectively. On effectiveness, the SwinV2-MoE model achieves superior accuracy in both pre-training and down-stream computer vision tasks such as COCO object detection than the counterpart dense model, indicating the readiness of Tutel for end-to-end real-world model training and inference. SwinV2-MoE is open sourced in https://github.com/microsoft/Swin-Transformer .

ModelDiff: Testing-Based DNN Similarity Comparison for Model Reuse Detection

The knowledge of a deep learning model may be transferred to a student model, leading to intellectual property infringement or vulnerability propagation. Detecting such knowledge reuse is nontrivial because the suspect models may not be white-box accessible and/or may serve different tasks. In this paper, we propose ModelDiff, a testing-based approach to deep learning model similarity comparison. Instead of directly comparing the weights, activations, or outputs of two models, we compare their behavioral patterns on the same set of test inputs. Specifically, the behavioral pattern of a model is represented as a decision distance vector (DDV), in which each element is the distance between the model's reactions to a pair of inputs. The knowledge similarity between two models is measured with the cosine similarity between their DDVs. To evaluate ModelDiff, we created a benchmark that contains 144 pairs of models that cover most popular model reuse methods, including transfer learning, model compression, and model stealing. Our method achieved 91.7% correctness on the benchmark, which demonstrates the effectiveness of using ModelDiff for model reuse detection. A study on mobile deep learning apps has shown the feasibility of ModelDiff on real-world models.