arXiv subjects

Huazhong Yang

Publications and source records attributed to Huazhong Yang.

At least 37 records · Page 2Linked to original sources

arXiv2024

Efficient Expert Pruning for Sparse Mixture-of-Experts Language Models: Enhancing Performance and Reducing Inference Costs

The rapid advancement of large language models (LLMs) has led to architectures with billions to trillions of parameters, posing significant deployment challenges due to their substantial demands on memory, processing power, and energy consumption. Sparse Mixture-of-Experts (SMoE) architectures have emerged as a solution, activating only a subset of parameters per token, thereby achieving faster inference while maintaining performance. However, SMoE models still face limitations in broader deployment due to their large parameter counts and significant GPU memory requirements. In this work, we introduce a gradient-free evolutionary strategy named EEP (Efficient Expert P}runing) to enhance the pruning of experts in SMoE models. EEP relies solely on model inference (i.e., no gradient computation) and achieves greater sparsity while maintaining or even improving performance on downstream tasks. EEP can be used to reduce both the total number of experts (thus saving GPU memory) and the number of active experts (thus accelerating inference). For example, we demonstrate that pruning up to 75% of experts in Mixtral $8\times7$B-Instruct results in a substantial reduction in parameters with minimal performance loss. Remarkably, we observe improved performance on certain tasks, such as a significant increase in accuracy on the SQuAD dataset (from 53.4% to 75.4%), when pruning half of the experts. With these results, EEP not only lowers the barrier to deploying SMoE models,but also challenges the conventional understanding of model pruning by showing that fewer experts can lead to better task-specific performance without any fine-tuning. Code is available at https://github.com/imagination-research/EEP.

cs.LG↗

arXiv2024

Evaluating Quantized Large Language Models

Post-training quantization (PTQ) has emerged as a promising technique to reduce the cost of large language models (LLMs). Specifically, PTQ can effectively mitigate memory consumption and reduce computational overhead in LLMs. To meet the requirements of both high efficiency and performance across diverse scenarios, a comprehensive evaluation of quantized LLMs is essential to guide the selection of quantization methods. This paper presents a thorough evaluation of these factors by evaluating the effect of PTQ on Weight, Activation, and KV Cache on 11 model families, including OPT, LLaMA2, Falcon, Bloomz, Mistral, ChatGLM, Vicuna, LongChat, StableLM, Gemma, and Mamba, with parameters ranging from 125M to 180B. The evaluation encompasses five types of tasks: basic NLP, emergent ability, trustworthiness, dialogue, and long-context tasks. Moreover, we also evaluate the state-of-the-art (SOTA) quantization methods to demonstrate their applicability. Based on the extensive experiments, we systematically summarize the effect of quantization, provide recommendations to apply quantization techniques, and point out future directions. The code can be found in https://github.com/thu-nics/qllm-eval.

cs.CL↗

arXiv2024

A Dual Curriculum Learning Framework for Multi-UAV Pursuit-Evasion in Diverse Environments

This paper addresses multi-UAV pursuit-evasion, where a group of drones cooperates to capture a fast evader in a confined environment with obstacles. Existing heuristic algorithms, which simplify the pursuit-evasion problem, often lack expressive coordination strategies and struggle to capture the evader in extreme scenarios, such as when the evader moves at high speeds. In contrast, reinforcement learning (RL) has been applied to this problem and has the potential to obtain highly cooperative capture strategies. However, RL-based methods face challenges in training for complex 3-dimensional scenarios with diverse task settings due to the vast exploration space. The dynamics constraints of drones further restrict the ability of reinforcement learning to acquire high-performance capture strategies. In this work, we introduce a dual curriculum learning framework, named DualCL, which addresses multi-UAV pursuit-evasion in diverse environments and demonstrates zero-shot transfer ability to unseen scenarios. DualCL comprises two main components: the Intrinsic Parameter Curriculum Proposer, which progressively suggests intrinsic parameters from easy to hard to improve the capture capability of drones, and the External Environment Generator, tasked with exploring unresolved scenarios and generating appropriate training distributions of external environment parameters. The simulation experimental results show that DualCL significantly outperforms baseline methods, achieving over 90% capture rate and reducing the capture timestep by at least 27.5% in the training scenarios. Additionally, it exhibits the best zero-shot generalization ability in unseen environments. Moreover, we demonstrate the transferability of our pursuit strategy from simulation to real-world environments. Further details can be found on the project website at https://sites.google.com/view/dualcl.

cs.LG↗

arXiv2024

A 65nm 8b-Activation 8b-Weight SRAM-Based Charge-Domain Computing-in-Memory Macro Using A Fully-Parallel Analog Adder Network and A Single-ADC Interface

Performing data-intensive tasks in the von Neumann architecture is challenging to achieve both high performance and power efficiency due to the memory wall bottleneck. Computing-in-memory (CiM) is a promising mitigation approach by enabling parallel in-situ multiply-accumulate (MAC) operations within the memory with support from the peripheral interface and datapath. SRAM-based charge-domain CiM (CD-CiM) has shown its potential of enhanced power efficiency and computing accuracy. However, existing SRAM-based CD-CiM faces scaling challenges to meet the throughput requirement of high-performance multi-bit-quantization applications. This paper presents an SRAM-based high-throughput ReLU-optimized CD-CiM macro. It is capable of completing MAC and ReLU of two signed 8b vectors in one CiM cycle with only one A/D conversion. Along with non-linearity compensation for the analog computing and A/D conversion interfaces, this work achieves 51.2GOPS throughput and 10.3TOPS/W energy efficiency, while showing 88.6% accuracy in the CIFAR-10 dataset.

cs.AR↗

arXiv2024

FlashEval: Towards Fast and Accurate Evaluation of Text-to-image Diffusion Generative Models

In recent years, there has been significant progress in the development of text-to-image generative models. Evaluating the quality of the generative models is one essential step in the development process. Unfortunately, the evaluation process could consume a significant amount of computational resources, making the required periodic evaluation of model performance (e.g., monitoring training progress) impractical. Therefore, we seek to improve the evaluation efficiency by selecting the representative subset of the text-image dataset. We systematically investigate the design choices, including the selection criteria (textural features or image-based metrics) and the selection granularity (prompt-level or set-level). We find that the insights from prior work on subset selection for training data do not generalize to this problem, and we propose FlashEval, an iterative search algorithm tailored to evaluation data selection. We demonstrate the effectiveness of FlashEval on ranking diffusion models with various configurations, including architectures, quantization levels, and sampler schedules on COCO and DiffusionDB datasets. Our searched 50-item subset could achieve comparable evaluation quality to the randomly sampled 500-item subset for COCO annotations on unseen models, achieving a 10x evaluation speedup. We release the condensed subset of these commonly used datasets to help facilitate diffusion algorithm design and evaluation, and open-source FlashEval as a tool for condensing future datasets, accessible at https://github.com/thu-nics/FlashEval.

cs.CV↗

arXiv2024

Skeleton-of-Thought: Prompting LLMs for Efficient Parallel Generation

This work aims at decreasing the end-to-end generation latency of large language models (LLMs). One of the major causes of the high generation latency is the sequential decoding approach adopted by almost all state-of-the-art LLMs. In this work, motivated by the thinking and writing process of humans, we propose Skeleton-of-Thought (SoT), which first guides LLMs to generate the skeleton of the answer, and then conducts parallel API calls or batched decoding to complete the contents of each skeleton point in parallel. Not only does SoT provide considerable speed-ups across 12 LLMs, but it can also potentially improve the answer quality on several question categories. SoT is an initial attempt at data-centric optimization for inference efficiency, and showcases the potential of eliciting high-quality answers by explicitly planning the answer structure in language.

cs.CL↗

arXiv2023

Accelerate Multi-Agent Reinforcement Learning in Zero-Sum Games with Subgame Curriculum Learning

Learning Nash equilibrium (NE) in complex zero-sum games with multi-agent reinforcement learning (MARL) can be extremely computationally expensive. Curriculum learning is an effective way to accelerate learning, but an under-explored dimension for generating a curriculum is the difficulty-to-learn of the subgames -- games induced by starting from a specific state. In this work, we present a novel subgame curriculum learning framework for zero-sum games. It adopts an adaptive initial state distribution by resetting agents to some previously visited states where they can quickly learn to improve performance. Building upon this framework, we derive a subgame selection metric that approximates the squared distance to NE values and further adopt a particle-based state sampler for subgame generation. Integrating these techniques leads to our new algorithm, Subgame Automatic Curriculum Learning (SACL), which is a realization of the subgame curriculum learning framework. SACL can be combined with any MARL algorithm such as MAPPO. Experiments in the particle-world environment and Google Research Football environment show SACL produces much stronger policies than baselines. In the challenging hide-and-seek quadrant environment, SACL produces all four emergent stages and uses only half the samples of MAPPO with self-play. The project website is at https://sites.google.com/view/sacl-rl.

cs.LG↗

arXiv2023

A Unified Sampling Framework for Solver Searching of Diffusion Probabilistic Models

Recent years have witnessed the rapid progress and broad application of diffusion probabilistic models (DPMs). Sampling from DPMs can be viewed as solving an ordinary differential equation (ODE). Despite the promising performance, the generation of DPMs usually consumes much time due to the large number of function evaluations (NFE). Though recent works have accelerated the sampling to around 20 steps with high-order solvers, the sample quality with less than 10 NFE can still be improved. In this paper, we propose a unified sampling framework (USF) to study the optional strategies for solver. Under this framework, we further reveal that taking different solving strategies at different timesteps may help further decrease the truncation error, and a carefully designed \emph{solver schedule} has the potential to improve the sample quality by a large margin. Therefore, we propose a new sampling framework based on the exponential integral formulation that allows free choices of solver strategy at each step and design specific decisions for the framework. Moreover, we propose $S^3$, a predictor-based search method that automatically optimizes the solver schedule to get a better time-quality trade-off of sampling. We demonstrate that $S^3$ can find outstanding solver schedules which outperform the state-of-the-art sampling methods on CIFAR-10, CelebA, ImageNet, and LSUN-Bedroom datasets. Specifically, we achieve 2.69 FID with 10 NFE and 6.86 FID with 5 NFE on CIFAR-10 dataset, outperforming the SOTA method significantly. We further apply $S^3$ to Stable-Diffusion model and get an acceleration ratio of 2$\times$, showing the feasibility of sampling in very few steps without retraining the neural network.

cs.AI↗

arXiv2023

Active Neural Topological Mapping for Multi-Agent Exploration

This paper investigates the multi-agent cooperative exploration problem, which requires multiple agents to explore an unseen environment via sensory signals in a limited time. A popular approach to exploration tasks is to combine active mapping with planning. Metric maps capture the details of the spatial representation, but are with high communication traffic and may vary significantly between scenarios, resulting in inferior generalization. Topological maps are a promising alternative as they consist only of nodes and edges with abstract but essential information and are less influenced by the scene structures. However, most existing topology-based exploration tasks utilize classical methods for planning, which are time-consuming and sub-optimal due to their handcrafted design. Deep reinforcement learning (DRL) has shown great potential for learning (near) optimal policies through fast end-to-end inference. In this paper, we propose Multi-Agent Neural Topological Mapping (MANTM) to improve exploration efficiency and generalization for multi-agent exploration tasks. MANTM mainly comprises a Topological Mapper and a novel RL-based Hierarchical Topological Planner (HTP). The Topological Mapper employs a visual encoder and distance-based heuristics to construct a graph containing main nodes and their corresponding ghost nodes. The HTP leverages graph neural networks to capture correlations between agents and graph nodes in a coarse-to-fine manner for effective global goal selection. Extensive experiments conducted in a physically-realistic simulator, Habitat, demonstrate that MANTM reduces the steps by at least 26.40% over planning-based baselines and by at least 7.63% over RL-based competitors in unseen scenarios.

cs.RO↗

arXiv2023

Ada3D : Exploiting the Spatial Redundancy with Adaptive Inference for Efficient 3D Object Detection

Voxel-based methods have achieved state-of-the-art performance for 3D object detection in autonomous driving. However, their significant computational and memory costs pose a challenge for their application to resource-constrained vehicles. One reason for this high resource consumption is the presence of a large number of redundant background points in Lidar point clouds, resulting in spatial redundancy in both 3D voxel and dense BEV map representations. To address this issue, we propose an adaptive inference framework called Ada3D, which focuses on exploiting the input-level spatial redundancy. Ada3D adaptively filters the redundant input, guided by a lightweight importance predictor and the unique properties of the Lidar point cloud. Additionally, we utilize the BEV features' intrinsic sparsity by introducing the Sparsity Preserving Batch Normalization. With Ada3D, we achieve 40% reduction for 3D voxels and decrease the density of 2D BEV feature maps from 100% to 20% without sacrificing accuracy. Ada3D reduces the model computational and memory cost by 5x, and achieves 1.52x/1.45x end-to-end GPU latency and 1.5x/4.5x GPU peak memory optimization for the 3D and 2D backbone respectively.

cs.CV↗

arXiv2023

OMS-DPM: Optimizing the Model Schedule for Diffusion Probabilistic Models

Diffusion probabilistic models (DPMs) are a new class of generative models that have achieved state-of-the-art generation quality in various domains. Despite the promise, one major drawback of DPMs is the slow generation speed due to the large number of neural network evaluations required in the generation process. In this paper, we reveal an overlooked dimension -- model schedule -- for optimizing the trade-off between generation quality and speed. More specifically, we observe that small models, though having worse generation quality when used alone, could outperform large models in certain generation steps. Therefore, unlike the traditional way of using a single model, using different models in different generation steps in a carefully designed \emph{model schedule} could potentially improve generation quality and speed \emph{simultaneously}. We design OMS-DPM, a predictor-based search algorithm, to optimize the model schedule given an arbitrary generation time budget and a set of pre-trained models. We demonstrate that OMS-DPM can find model schedules that improve generation quality and speed than prior state-of-the-art methods across CIFAR-10, CelebA, ImageNet, and LSUN datasets. When applied to the public checkpoints of the Stable Diffusion model, we are able to accelerate the sampling by 2$\times$ while maintaining the generation quality.

cs.LG↗

arXiv2023

Asynchronous Multi-Agent Reinforcement Learning for Efficient Real-Time Multi-Robot Cooperative Exploration

We consider the problem of cooperative exploration where multiple robots need to cooperatively explore an unknown region as fast as possible. Multi-agent reinforcement learning (MARL) has recently become a trending paradigm for solving this challenge. However, existing MARL-based methods adopt action-making steps as the metric for exploration efficiency by assuming all the agents are acting in a fully synchronous manner: i.e., every single agent produces an action simultaneously and every single action is executed instantaneously at each time step. Despite its mathematical simplicity, such a synchronous MARL formulation can be problematic for real-world robotic applications. It can be typical that different robots may take slightly different wall-clock times to accomplish an atomic action or even periodically get lost due to hardware issues. Simply waiting for every robot being ready for the next action can be particularly time-inefficient. Therefore, we propose an asynchronous MARL solution, Asynchronous Coordination Explorer (ACE), to tackle this real-world challenge. We first extend a classical MARL algorithm, multi-agent PPO (MAPPO), to the asynchronous setting and additionally apply action-delay randomization to enforce the learned policy to generalize better to varying action delays in the real world. Moreover, each navigation agent is represented as a team-size-invariant CNN-based policy, which greatly benefits real-robot deployment by handling possible robot lost and allows bandwidth-efficient intra-agent communication through low-dimensional CNN features. We first validate our approach in a grid-based scenario. Both simulation and real-robot results show that ACE reduces over 10% actual exploration time compared with classical approaches. We also apply our framework to a high-fidelity visual-based environment, Habitat, achieving 28% improvement in exploration efficiency.

cs.RO↗

arXiv2023

ASMCap: An Approximate String Matching Accelerator for Genome Sequence Analysis Based on Capacitive Content Addressable Memory

Genome sequence analysis is a powerful tool in medical and scientific research. Considering the inevitable sequencing errors and genetic variations, approximate string matching (ASM) has been adopted in practice for genome sequencing. However, with exponentially increasing bio-data, ASM hardware acceleration is facing severe challenges in improving the throughput and energy efficiency with the accuracy constraint. This paper presents ASMCap, an ASM acceleration approach for genome sequence analysis with hardware-algorithm co-optimization. At the circuit level, ASMCap adopts charge-domain computing based on the capacitive multi-level content addressable memories (ML-CAMs), and outperforms the state-of-the-art ML-CAM-based ASM accelerators EDAM with higher accuracy and energy efficiency. ASMCap also has misjudgment correction capability with two proposed hardware-friendly strategies, namely the Hamming-Distance Aid Correction (HDAC) for the substitution-dominant edits and the Threshold-Aware Sequence Rotation (TASR) for the consecutive indels. Evaluation results show that ASMCap can achieve an average of 1.2x (from 74.7% to 87.6%) and up to 1.8x (from 46.3% to 81.2%) higher F1 score (the key metric of accuracy), 1.4x speedup, and 10.8x energy efficiency improvement compared with EDAM. Compared with the other ASM accelerators, including ResMA based on the comparison matrix, and SaVI based on the seeding strategy, ASMCap achieves an average improvement of 174x and 61x speedup, and 8.7e3x and 943x higher energy efficiency, respectively.

cs.AR↗

arXiv2023

Learning Graph-Enhanced Commander-Executor for Multi-Agent Navigation

This paper investigates the multi-agent navigation problem, which requires multiple agents to reach the target goals in a limited time. Multi-agent reinforcement learning (MARL) has shown promising results for solving this issue. However, it is inefficient for MARL to directly explore the (nearly) optimal policy in the large search space, which is exacerbated as the agent number increases (e.g., 10+ agents) or the environment is more complex (e.g., 3D simulator). Goal-conditioned hierarchical reinforcement learning (HRL) provides a promising direction to tackle this challenge by introducing a hierarchical structure to decompose the search space, where the low-level policy predicts primitive actions in the guidance of the goals derived from the high-level policy. In this paper, we propose Multi-Agent Graph-Enhanced Commander-Executor (MAGE-X), a graph-based goal-conditioned hierarchical method for multi-agent navigation tasks. MAGE-X comprises a high-level Goal Commander and a low-level Action Executor. The Goal Commander predicts the probability distribution of goals and leverages them to assign each agent the most appropriate final target. The Action Executor utilizes graph neural networks (GNN) to construct a subgraph for each agent that only contains crucial partners to improve cooperation. Additionally, the Goal Encoder in the Action Executor captures the relationship between the agent and the designated goal to encourage the agent to reach the final target. The results show that MAGE-X outperforms the state-of-the-art MARL baselines with a 100% success rate with only 3 million training steps in multi-agent particle environments (MPE) with 50 agents, and at least a 12% higher success rate and 2x higher data efficiency in a more complicated quadrotor 3D navigation task.

cs.RO↗

arXiv2023

Eliminating Leakage in Volatile Memory with Anti-Ferroelectric Transistors

Cache serves as a temporary data memory module in many general-purpose processors and domain-specific accelerators. Its density, power, speed, and reliability play a critical role in enhancing the overall system performance and quality of service. Conventional volatile memories, including static random-access memory (SRAM) and embedded dynamic random-access memory (eDRAM) in the complementary metal-oxide-semiconductor technology, have high performance and good reliability. However, the inherent leakage in both SRAM and eDRAM hinders further improvement towards smaller feature sizes and higher energy efficiency. Although the emerging nonvolatile memories can eliminate the leakage efficiently, the penalties of lower speed and degraded reliability are significant. This article reveals a new opportunity towards leakage-free volatile static memory beyond the known paradigms of existing volatile and nonvolatile memories. By engineering a double-well energy landscape with the assistance of a clamping voltage bias, leakage-free and refresh-free state retention of volatile memory is achieved for the first time. This new memory is highlighted by both the ultra-low leakage of nonvolatile memories and the speed, energy, and reliability advantages of volatile memories. A proof-of-concept memory is demonstrated using in-house anti-ferroelectric field-effect transistors (AFeFETs), delivering an extrapolated endurance of about 1012 cycles, a retention time of over 10 years, and no subthreshold channel leakage current. Such a new concept of AFeFET-based memory enables an improved balance between density, power, and reliability beyond all existing memory solutions.

cs.ET↗

arXiv2022

Learning Efficient Multi-Agent Cooperative Visual Exploration

We tackle the problem of cooperative visual exploration where multiple agents need to jointly explore unseen regions as fast as possible based on visual signals. Classical planning-based methods often suffer from expensive computation overhead at each step and a limited expressiveness of complex cooperation strategy. By contrast, reinforcement learning (RL) has recently become a popular paradigm for tackling this challenge due to its modeling capability of arbitrarily complex strategies and minimal inference overhead. In this paper, we extend the state-of-the-art single-agent visual navigation method, Active Neural SLAM (ANS), to the multi-agent setting by introducing a novel RL-based planning module, Multi-agent Spatial Planner (MSP).MSP leverages a transformer-based architecture, Spatial-TeamFormer, which effectively captures spatial relations and intra-agent interactions via hierarchical spatial self-attentions. In addition, we also implement a few multi-agent enhancements to process local information from each agent for an aligned spatial representation and more precise planning. Finally, we perform policy distillation to extract a meta policy to significantly improve the generalization capability of final policy. We call this overall solution, Multi-Agent Active Neural SLAM (MAANS). MAANS substantially outperforms classical planning-based baselines for the first time in a photo-realistic 3D simulator, Habitat. Code and videos can be found at https://sites.google.com/view/maans.

cs.CV↗

arXiv2022

Block-Wise Dynamic-Precision Neural Network Training Acceleration via Online Quantization Sensitivity Analytics

Data quantization is an effective method to accelerate neural network training and reduce power consumption. However, it is challenging to perform low-bit quantized training: the conventional equal-precision quantization will lead to either high accuracy loss or limited bit-width reduction, while existing mixed-precision methods offer high compression potential but failed to perform accurate and efficient bit-width assignment. In this work, we propose DYNASTY, a block-wise dynamic-precision neural network training framework. DYNASTY provides accurate data sensitivity information through fast online analytics, and maintains stable training convergence with an adaptive bit-width map generator. Network training experiments on CIFAR-100 and ImageNet dataset are carried out, and compared to 8-bit quantization baseline, DYNASTY brings up to $5.1\times$ speedup and $4.7\times$ energy consumption reduction with no accuracy drop and negligible hardware overhead.

cs.LG↗

arXiv2022

Cross-layer Attention Network for Fine-grained Visual Categorization

Learning discriminative representations for subtle localized details plays a significant role in Fine-grained Visual Categorization (FGVC). Compared to previous attention-based works, our work does not explicitly define or localize the part regions of interest; instead, we leverage the complementary properties of different stages of the network, and build a mutual refinement mechanism between the mid-level feature maps and the top-level feature map by our proposed Cross-layer Attention Network (CLAN). Specifically, CLAN is composed of 1) the Cross-layer Context Attention (CLCA) module, which enhances the global context information in the intermediate feature maps with the help of the top-level feature map, thereby improving the expressive power of the middle layers, and 2) the Cross-layer Spatial Attention (CLSA) module, which takes advantage of the local attention in the mid-level feature maps to boost the feature extraction of local regions at the top-level feature maps. Experimental results show our approach achieves state-of-the-art on three publicly available fine-grained recognition datasets (CUB-200-2011, Stanford Cars and FGVC-Aircraft). Ablation studies and visualizations are provided to understand our approach. Experimental results show our approach achieves state-of-the-art on three publicly available fine-grained recognition datasets (CUB-200-2011, Stanford Cars and FGVC-Aircraft).

cs.CV↗