arXiv subjects

Bin Cui

Publications and source records attributed to Bin Cui.

At least 163 records · Page 9Linked to original sources

arXiv2023

Angel-PTM: A Scalable and Economical Large-scale Pre-training System in Tencent

Recent years have witnessed the unprecedented achievements of large-scale pre-trained models, especially the Transformer models. Many products and services in Tencent Inc., such as WeChat, QQ, and Tencent Advertisement, have been opted in to gain the power of pre-trained models. In this work, we present Angel-PTM, a productive deep learning system designed for pre-training and fine-tuning Transformer models. Angel-PTM can train extremely large-scale models with hierarchical memory efficiently. The key designs of Angel-PTM are the fine-grained memory management via the Page abstraction and a unified scheduling method that coordinate the computations, data movements, and communications. Furthermore, Angel-PTM supports extreme model scaling with SSD storage and implements the lock-free updating mechanism to address the SSD I/O bandwidth bottlenecks. Experimental results demonstrate that Angel-PTM outperforms existing systems by up to 114.8% in terms of maximum model scale as well as up to 88.9% in terms of training throughput. Additionally, experiments on GPT3-175B and T5-MoE-1.2T models utilizing hundreds of GPUs verify the strong scalability of Angel-PTM.

cs.LG↗

arXiv2023

Transfer Learning for Bayesian Optimization: A Survey

A wide spectrum of design and decision problems, including parameter tuning, A/B testing and drug design, intrinsically are instances of black-box optimization. Bayesian optimization (BO) is a powerful tool that models and optimizes such expensive "black-box" functions. However, at the beginning of optimization, vanilla Bayesian optimization methods often suffer from slow convergence issue due to inaccurate modeling based on few trials. To address this issue, researchers in the BO community propose to incorporate the spirit of transfer learning to accelerate optimization process, which could borrow strength from the past tasks (source tasks) to accelerate the current optimization problem (target task). This survey paper first summarizes transfer learning methods for Bayesian optimization from four perspectives: initial points design, search space design, surrogate model, and acquisition function. Then it highlights its methodological aspects and technical details for each approach. Finally, it showcases a wide range of applications and proposes promising future directions.

cs.LG↗

arXiv2023

DivBO: Diversity-aware CASH for Ensemble Learning

The Combined Algorithm Selection and Hyperparameters optimization (CASH) problem is one of the fundamental problems in Automated Machine Learning (AutoML). Motivated by the success of ensemble learning, recent AutoML systems build post-hoc ensembles to output the final predictions instead of using the best single learner. However, while most CASH methods focus on searching for a single learner with the best performance, they neglect the diversity among base learners (i.e., they may suggest similar configurations to previously evaluated ones), which is also a crucial consideration when building an ensemble. To tackle this issue and further enhance the ensemble performance, we propose DivBO, a diversity-aware framework to inject explicit search of diversity into the CASH problems. In the framework, we propose to use a diversity surrogate to predict the pair-wise diversity of two unseen configurations. Furthermore, we introduce a temporary pool and a weighted acquisition function to guide the search of both performance and diversity based on Bayesian optimization. Empirical results on 15 public datasets show that DivBO achieves the best average ranks (1.82 and 1.73) on both validation and test errors among 10 compared methods, including post-hoc designs in recent AutoML systems and state-of-the-art baselines for ensemble learning on CASH problems.

cs.LG↗

arXiv2023

Switchable Giant Bulk Photocurrents and Photo-spin-currents in Monolayer PT-symmetric Anti-ferromagnet MnPSe3

Converting light into steady currents and spin-currents in two-dimensional (2D) platform is essential for future energy harvesting and spintronics. We show that the giant and modulable bulk photovoltaic effects (BPVEs) can be achieved in air-stable 2D antiferromagnet (AFM) monolayer MnPSe3, with nonlinear photoconductance > 4000 nm$\cdotμ$A/V2 and photo-spin-conductance > 2000 (nm$\cdotμ$A/V2 $\hbar$/2e) in the visible spectrum. The propagation and the spin-polarizations of photocurrents can be switched via simply rotating the N$é$el vector. We unveil that the PT-symmetry, mirror symmetries, and spin-orbital-couplings are the keys for the observed sizable and controllable 2D BPVEs. All the results provide insights into the BPVEs of 2D AFM, and suggest that the layered MnPSe3 is an outstanding 2D platform for energy device and photo-spintronics.

cond-mat.mtrl-sci↗

arXiv2023

Layer-number-dependent spin Hall effects in transition metal monocarbides $M_{2}\rm{C}$ ($M=\rm{V}, \rm{Nb}, \rm{Ta}$)

The recent discovery of strong spin Hall effects (SHE) in 2D layered topological semimetals has attracted intensive attention due to its exotic electronic properties and potential applications in spintronic devices. In this paper, we systematically study the topological properties and intrinsic SHE of layered transition metal carbides $M_{2}\rm{C}$ ($M=\rm{V}, \rm{Nb}, \rm{Ta}$). The results show that both bulk and monolayer $M_{2}\rm{C}$ have symmetry-protected nodal points (NPs) and lines (NLs) originating from the $d$ band crossing near the Fermi level ($E_F$). The inclusion of SOC breaks the degeneracy of NLs and NPs, contributing to large spin Hall conductivity (SHC) up to $\sim$1100 and $\sim$200 $(\hbar / e)(Ω\mathrm{cm})^{-1}$ for bulk and monolayer Ta$_{2}$C, respectively. Remarkably, we find that magnitude of SHC exhibits a significant enhancement by increasing the layer number. For eight-layer Ta$_{2}$C, the maximum value of SHC can reach up to $\sim$600 $(\hbar / e)(Ω\mathrm{cm})^{-1}$, comparable to many reported 3D topological materials. Analysis of spin Berry curvature reveals that the large SHC originates from layer-number-dependent nodal line structure near the $E_F$, in which the repeated crossover between valence and conduction bands creates large amounts of NPs along the $Γ\rm{-K}$ route. Our findings not only provide a new platform for experimental research of low-dimensional SHE, but also suggest an effective way of realizing giant SHE by controlling layer thickness.

cond-mat.mtrl-sci↗

arXiv2023

AutoML for Deep Recommender Systems: A Survey

Recommender systems play a significant role in information filtering and have been utilized in different scenarios, such as e-commerce and social media. With the prosperity of deep learning, deep recommender systems show superior performance by capturing non-linear information and item-user relationships. However, the design of deep recommender systems heavily relies on human experiences and expert knowledge. To tackle this problem, Automated Machine Learning (AutoML) is introduced to automatically search for the proper candidates for different parts of deep recommender systems. This survey performs a comprehensive review of the literature in this field. Firstly, we propose an abstract concept for AutoML for deep recommender systems (AutoRecSys) that describes its building blocks and distinguishes it from conventional AutoML techniques and recommender systems. Secondly, we present a taxonomy as a classification framework containing feature selection search, embedding dimension search, feature interaction search, model architecture search, and other components search. Furthermore, we put a particular emphasis on the search space and search strategy, as they are the common thread to connect all methods within each category and enable practitioners to analyze and compare various approaches. Finally, we propose four future promising research directions that will lead this line of research.

cs.IR↗

arXiv2022

Efficient Graph Neural Network Inference at Large Scale

Graph neural networks (GNNs) have demonstrated excellent performance in a wide range of applications. However, the enormous size of large-scale graphs hinders their applications under real-time inference scenarios. Although existing scalable GNNs leverage linear propagation to preprocess the features and accelerate the training and inference procedure, these methods still suffer from scalability issues when making inferences on unseen nodes, as the feature preprocessing requires the graph is known and fixed. To speed up the inference in the inductive setting, we propose a novel adaptive propagation order approach that generates the personalized propagation order for each node based on its topological information. This could successfully avoid the redundant computation of feature propagation. Moreover, the trade-off between accuracy and inference latency can be flexibly controlled by simple hyper-parameters to match different latency constraints of application scenarios. To compensate for the potential inference accuracy loss, we further propose Inception Distillation to exploit the multi scale reception information and improve the inference performance. Extensive experiments are conducted on four public datasets with different scales and characteristics, and the experimental results show that our proposed inference acceleration framework outperforms the SOTA graph inference acceleration baselines in terms of both accuracy and efficiency. In particular, the advantage of our proposed method is more significant on larger-scale datasets, and our framework achieves $75\times$ inference speedup on the largest Ogbn-products dataset.

cs.LG↗

arXiv2022

CALIP: Zero-Shot Enhancement of CLIP with Parameter-free Attention

Contrastive Language-Image Pre-training (CLIP) has been shown to learn visual representations with great transferability, which achieves promising accuracy for zero-shot classification. To further improve its downstream performance, existing works propose additional learnable modules upon CLIP and fine-tune them by few-shot training sets. However, the resulting extra training cost and data requirement severely hinder the efficiency for model deployment and knowledge transfer. In this paper, we introduce a free-lunch enhancement method, CALIP, to boost CLIP's zero-shot performance via a parameter-free Attention module. Specifically, we guide visual and textual representations to interact with each other and explore cross-modal informative features via attention. As the pre-training has largely reduced the embedding distances between two modalities, we discard all learnable parameters in the attention and bidirectionally update the multi-modal features, enabling the whole process to be parameter-free and training-free. In this way, the images are blended with textual-aware signals and the text representations become visual-guided for better adaptive zero-shot alignment. We evaluate CALIP on various benchmarks of 14 datasets for both 2D image and 3D point cloud few-shot classification, showing consistent zero-shot performance improvement over CLIP. Based on that, we further insert a small number of linear layers in CALIP's attention module and verify our robustness under the few-shot settings, which also achieves leading performance compared to existing methods. Those extensive experiments demonstrate the superiority of our approach for efficient enhancement of CLIP.

cs.CV↗

arXiv2022

Galvatron: Efficient Transformer Training over Multiple GPUs Using Automatic Parallelism

Transformer models have achieved state-of-the-art performance on various domains of applications and gradually becomes the foundations of the advanced large deep learning (DL) models. However, how to train these models over multiple GPUs efficiently is still challenging due to a large number of parallelism choices. Existing DL systems either rely on manual efforts to make distributed training plans or apply parallelism combinations within a very limited search space. In this approach, we propose Galvatron, a new system framework that incorporates multiple popular parallelism dimensions and automatically finds the most efficient hybrid parallelism strategy. To better explore such a rarely huge search space, we 1) involve a decision tree to make decomposition and pruning based on some reasonable intuitions, and then 2) design a dynamic programming search algorithm to generate the optimal plan. Evaluations on four representative Transformer workloads show that Galvatron could perform automatically distributed training with different GPU memory budgets. Among all evluated scenarios, Galvatron always achieves superior system throughput compared to previous work with limited parallelism.

cs.LG↗

arXiv2022

Diffusion-Based Scene Graph to Image Generation with Masked Contrastive Pre-Training

Generating images from graph-structured inputs, such as scene graphs, is uniquely challenging due to the difficulty of aligning nodes and connections in graphs with objects and their relations in images. Most existing methods address this challenge by using scene layouts, which are image-like representations of scene graphs designed to capture the coarse structures of scene images. Because scene layouts are manually crafted, the alignment with images may not be fully optimized, causing suboptimal compliance between the generated images and the original scene graphs. To tackle this issue, we propose to learn scene graph embeddings by directly optimizing their alignment with images. Specifically, we pre-train an encoder to extract both global and local information from scene graphs that are predictive of the corresponding images, relying on two loss functions: masked autoencoding loss and contrastive loss. The former trains embeddings by reconstructing randomly masked image regions, while the latter trains embeddings to discriminate between compliant and non-compliant images according to the scene graph. Given these embeddings, we build a latent diffusion model to generate images from scene graphs. The resulting method, called SGDiff, allows for the semantic manipulation of generated images by modifying scene graph nodes and connections. On the Visual Genome and COCO-Stuff datasets, we demonstrate that SGDiff outperforms state-of-the-art methods, as measured by both the Inception Score and Fréchet Inception Distance (FID) metrics. We will release our source code and trained models at https://github.com/YangLing0818/SGDiff.

cs.CV↗

arXiv2022

HetuMoE: An Efficient Trillion-scale Mixture-of-Expert Distributed Training System

As giant dense models advance quality but require large amounts of GPU budgets for training, the sparsely gated Mixture-of-Experts (MoE), a kind of conditional computation architecture, is proposed to scale models while keeping their computation constant. Specifically, the input tokens are routed by the gate network and only activates part of the expert network. Existing MoE training systems only support part of mainstream MoE models (e.g. Top k) training under expensive high-bandwidth GPU clusters. In this paper, we present HetuMoE, a high-performance large-scale sparse MoE training system built on Hetu. HetuMoE provides multiple gating strategies and efficient GPU kernel implementations. To further improve the training efficiency on commodity GPU clusters (e.g, with only 1 NiC), we introduce the hierarchical AllToAll communication that combines hierarchical networks and aggregating messages. Compared with existing state-of-the-art MoE systems, HetuMoE obtains at least 15% speedup. Specifically, HetuMoE outperforms DeepSpeed-MoE up to 8.1x under the switch gate with a batch size of 32. Our code is available at: https://github.com/PKU-DAIR/Hetu.

cs.DC↗

arXiv2022

EvoMoE: An Evolutional Mixture-of-Experts Training Framework via Dense-To-Sparse Gate

Mixture-of-experts (MoE) is becoming popular due to its success in improving the model quality, especially in Transformers. By routing tokens with a sparse gate to a few experts (i.e., a small pieces of the full model), MoE can easily increase the model parameters to a very large scale while keeping the computation cost in a constant level. Most existing works just initialize some random experts, set a fixed gating strategy (e.g., Top-k), and train the model from scratch in an ad-hoc way. We identify that these MoE models are suffering from the immature experts and unstable sparse gate, which are harmful to the convergence performance. In this paper, we propose an efficient end-to-end MoE training framework called EvoMoE. EvoMoE starts from training one single expert and gradually evolves into a large and sparse MoE structure. EvoMoE mainly contains two phases: the expert-diversify phase to train the base expert for a while and spawn multiple diverse experts from it, and the gate-sparsify phase to learn an adaptive sparse gate and activate a dynamic number of experts. EvoMoE naturally decouples the joint learning of both the experts and the sparse gate and focuses on learning the basic knowledge with a single expert at the early training stage. Then it diversifies the experts and continues to train the MoE with a novel Dense-to-Sparse gate (DTS-Gate). Specifically, instead of using a permanent sparse gate, DTS-Gate begins as a dense gate that routes tokens to all experts, then gradually and adaptively becomes sparser while routes to fewer experts. Evaluations are conducted on three popular models and tasks, including RoBERTa for masked language modeling task, GPT for language modeling task and Transformer for machine translation task. The results show that EvoMoE outperforms existing baselines, including Switch, BASE Layer, Hash Layer and StableMoE.

cs.LG↗

arXiv2022

Towards Communication-efficient Vertical Federated Learning Training via Cache-enabled Local Updates

Vertical federated learning (VFL) is an emerging paradigm that allows different parties (e.g., organizations or enterprises) to collaboratively build machine learning models with privacy protection. In the training phase, VFL only exchanges the intermediate statistics, i.e., forward activations and backward derivatives, across parties to compute model gradients. Nevertheless, due to its geo-distributed nature, VFL training usually suffers from the low WAN bandwidth. In this paper, we introduce CELU-VFL, a novel and efficient VFL training framework that exploits the local update technique to reduce the cross-party communication rounds. CELU-VFL caches the stale statistics and reuses them to estimate model gradients without exchanging the ad hoc statistics. Significant techniques are proposed to improve the convergence performance. First, to handle the stochastic variance problem, we propose a uniform sampling strategy to fairly choose the stale statistics for local updates. Second, to harness the errors brought by the staleness, we devise an instance weighting mechanism that measures the reliability of the estimated gradients. Theoretical analysis proves that CELU-VFL achieves a similar sub-linear convergence rate as vanilla VFL training but requires much fewer communication rounds. Empirical results on both public and real-world workloads validate that CELU-VFL can be up to six times faster than the existing works.

cs.LG↗

arXiv2022

Efficient End-to-End AutoML via Scalable Search Space Decomposition

End-to-end AutoML has attracted intensive interests from both academia and industry which automatically searches for ML pipelines in a space induced by feature engineering, algorithm/model selection, and hyper-parameter tuning. Existing AutoML systems, however, suffer from scalability issues when applying to application domains with large, high-dimensional search spaces. We present VolcanoML, a scalable and extensible framework that facilitates systematic exploration of large AutoML search spaces. VolcanoML introduces and implements basic building blocks that decompose a large search space into smaller ones, and allows users to utilize these building blocks to compose an execution plan for the AutoML problem at hand. VolcanoML further supports a Volcano-style execution model -- akin to the one supported by modern database systems -- to execute the plan constructed. Our evaluation demonstrates that, not only does VolcanoML raise the level of expressiveness for search space decomposition in AutoML, it also leads to actual findings of decomposition strategies that are significantly more efficient than the ones employed by state-of-the-art AutoML systems such as auto-sklearn. This paper is the extended version of the initial VolcanoML paper appeared in VLDB 2021.

cs.LG↗

arXiv2022

DFG-NAS: Deep and Flexible Graph Neural Architecture Search

Graph neural networks (GNNs) have been intensively applied to various graph-based applications. Despite their success, manually designing the well-behaved GNNs requires immense human expertise. And thus it is inefficient to discover the potentially optimal data-specific GNN architecture. This paper proposes DFG-NAS, a new neural architecture search (NAS) method that enables the automatic search of very deep and flexible GNN architectures. Unlike most existing methods that focus on micro-architectures, DFG-NAS highlights another level of design: the search for macro-architectures on how atomic propagation (\textbf{\texttt{P}}) and transformation (\textbf{\texttt{T}}) operations are integrated and organized into a GNN. To this end, DFG-NAS proposes a novel search space for \textbf{\texttt{P-T}} permutations and combinations based on message-passing dis-aggregation, defines four custom-designed macro-architecture mutations, and employs the evolutionary algorithm to conduct an efficient and effective search. Empirical studies on four node classification tasks demonstrate that DFG-NAS outperforms state-of-the-art manual designs and NAS methods of GNNs.

cs.LG↗

arXiv2022

NAFS: A Simple yet Tough-to-beat Baseline for Graph Representation Learning

Recently, graph neural networks (GNNs) have shown prominent performance in graph representation learning by leveraging knowledge from both graph structure and node features. However, most of them have two major limitations. First, GNNs can learn higher-order structural information by stacking more layers but can not deal with large depth due to the over-smoothing issue. Second, it is not easy to apply these methods on large graphs due to the expensive computation cost and high memory usage. In this paper, we present node-adaptive feature smoothing (NAFS), a simple non-parametric method that constructs node representations without parameter learning. NAFS first extracts the features of each node with its neighbors of different hops by feature smoothing, and then adaptively combines the smoothed features. Besides, the constructed node representation can further be enhanced by the ensemble of smoothed features extracted via different smoothing strategies. We conduct experiments on four benchmark datasets on two different application scenarios: node clustering and link prediction. Remarkably, NAFS with feature ensemble outperforms the state-of-the-art GNNs on these tasks and mitigates the aforementioned two limitations of most learning-based GNN counterparts.

cs.LG↗

arXiv2022

BlindFL: Vertical Federated Machine Learning without Peeking into Your Data

Due to the rising concerns on privacy protection, how to build machine learning (ML) models over different data sources with security guarantees is gaining more popularity. Vertical federated learning (VFL) describes such a case where ML models are built upon the private data of different participated parties that own disjoint features for the same set of instances, which fits many real-world collaborative tasks. Nevertheless, we find that existing solutions for VFL either support limited kinds of input features or suffer from potential data leakage during the federated execution. To this end, this paper aims to investigate both the functionality and security of ML modes in the VFL scenario. To be specific, we introduce BlindFL, a novel framework for VFL training and inference. First, to address the functionality of VFL models, we propose the federated source layers to unite the data from different parties. Various kinds of features can be supported efficiently by the federated source layers, including dense, sparse, numerical, and categorical features. Second, we carefully analyze the security during the federated execution and formalize the privacy requirements. Based on the analysis, we devise secure and accurate algorithm protocols, and further prove the security guarantees under the ideal-real simulation paradigm. Extensive experiments show that BlindFL supports diverse datasets and models efficiently whilst achieves robust privacy guarantees.

cs.LG↗

arXiv2022

Graph Attention MLP with Reliable Label Utilization

Graph neural networks (GNNs) have recently achieved state-of-the-art performance in many graph-based applications. Despite the high expressive power, they typically need to perform an expensive recursive neighborhood expansion in multiple training epochs and face a scalability issue. Moreover, most of them are inflexible since they are restricted to fixed-hop neighborhoods and insensitive to actual receptive field demands for different nodes. We circumvent these limitations by introducing a scalable and flexible Graph Attention Multilayer Perceptron (GAMLP). With the separation of the non-linear transformation and feature propagation, GAMLP significantly improves the scalability and efficiency by performing the propagation procedure in a pre-compute manner. With three principled receptive field attention, each node in GAMLP is flexible and adaptive in leveraging the propagated features over the different sizes of reception field. We conduct extensive evaluations on the three large open graph benchmarks (e.g., ogbn-papers100M, ogbn-products and ogbn-mag), demonstrating that GAMLP not only achieves the state-of-art performance, but also additionally provide high scalability and efficiency.

cs.LG↗