We introduce ZOOMER, a system deployed at Taobao, the largest e-commerce platform in China, for training and serving GNN-based recommendations over web-scale graphs. ZOOMER is designed for tackling two challenges presented by the massive user data at Taobao: low training/serving efficiency due to the huge scale of the graphs, and low recommendation quality due to the information overload which distracts the recommendation model from specific user intentions. ZOOMER achieves this by introducing a key concept, Region of Interests (ROI) in GNNs for recommendations, i.e., a neighborhood region in the graph with significant relevance to a strong user intention. ZOOMER narrows the focus from the whole graph and "zooms in" on the more relevant ROIs, thereby reducing the training/serving cost and mitigating the information overload at the same time. With carefully designed mechanisms, ZOOMER identifies the interest expressed by each recommendation request, constructs an ROI subgraph by sampling with respect to the interest, and guides the GNN to reweigh different parts of the ROI towards the interest by a multi-level attention module. Deployed as a large-scale distributed system, ZOOMER supports graphs with billions of nodes for training and thousands of requests per second for serving. ZOOMER achieves up to 14x speedup when downsizing sampling scales with comparable (even better) AUC performance than baseline methods. Besides, both the offline evaluation and online A/B test demonstrate the effectiveness of ZOOMER.
Graph Neural Networks (GNNs) have achieved great success in various tasks, but their performance highly relies on a large number of labeled nodes, which typically requires considerable human effort. GNN-based Active Learning (AL) methods are proposed to improve the labeling efficiency by selecting the most valuable nodes to label. Existing methods assume an oracle can correctly categorize all the selected nodes and thus just focus on the node selection. However, such an exact labeling task is costly, especially when the categorization is out of the domain of individual expert (oracle). The paper goes further, presenting a soft-label approach to AL on GNNs. Our key innovations are: i) relaxed queries where a domain expert (oracle) only judges the correctness of the predicted labels (a binary question) rather than identifying the exact class (a multi-class question), and ii) new criteria of maximizing information gain propagation for active learner with relaxed queries and soft labels. Empirical studies on public datasets demonstrate that our method significantly outperforms the state-of-the-art GNN-based AL methods in terms of both accuracy and labeling cost.
Graph neural networks (GNNs) have achieved state-of-the-art performance in various graph-based tasks. However, as mainstream GNNs are designed based on the neural message passing mechanism, they do not scale well to data size and message passing steps. Although there has been an emerging interest in the design of scalable GNNs, current researches focus on specific GNN design, rather than the general design space, limiting the discovery of potential scalable GNN models. This paper proposes PasCa, a new paradigm and system that offers a principled approach to systemically construct and explore the design space for scalable GNNs, rather than studying individual designs. Through deconstructing the message passing mechanism, PasCa presents a novel Scalable Graph Neural Architecture Paradigm (SGAP), together with a general architecture design space consisting of 150k different designs. Following the paradigm, we implement an auto-search engine that can automatically search well-performing and scalable GNN architectures to balance the trade-off between multiple criteria (e.g., accuracy and efficiency) via multi-objective optimization. Empirical studies on ten benchmark datasets demonstrate that the representative instances (i.e., PasCa-V1, V2, and V3) discovered by our system achieve consistent performance among competitive baselines. Concretely, PasCa-V3 outperforms the state-of-the-art GNN method JK-Net by 0.4\% in terms of predictive accuracy on our large industry dataset while achieving up to $28.3\times$ training speedups.
The ever-growing demand and complexity of machine learning are putting pressure on hyper-parameter tuning systems: while the evaluation cost of models continues to increase, the scalability of state-of-the-arts starts to become a crucial bottleneck. In this paper, inspired by our experience when deploying hyper-parameter tuning in a real-world application in production and the limitations of existing systems, we propose Hyper-Tune, an efficient and robust distributed hyper-parameter tuning framework. Compared with existing systems, Hyper-Tune highlights multiple system optimizations, including (1) automatic resource allocation, (2) asynchronous scheduling, and (3) multi-fidelity optimizer. We conduct extensive evaluations on benchmark datasets and a large-scale real-world dataset in production. Empirically, with the aid of these optimizations, Hyper-Tune outperforms competitive hyper-parameter tuning systems on a wide range of scenarios, including XGBoost, CNN, RNN, and some architectural hyper-parameters for neural networks. Compared with the state-of-the-art BOHB and A-BOHB, Hyper-Tune achieves up to 11.2x and 5.1x speedups, respectively.
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 that each only contains part of the full model, MoE keeps the model size unchanged and significantly reduces per-token computation, which effectively scales neural networks. However, we found that the current approach of jointly training experts and the sparse gate introduces a negative impact on model accuracy, diminishing the efficiency of expensive large-scale model training. In this work, we proposed Dense-To-Sparse gate (DTS-Gate) for MoE training. 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. MoE with DTS-Gate naturally decouples the training of experts and the sparse gate by training all experts at first and then learning the sparse gate. Experiments show that compared with the state-of-the-art Switch-Gate in GPT-MoE(1.5B) model with OpenWebText dataset(40GB), DTS-Gate can obtain 2.0x speed-up to reach the same validation perplexity, as well as higher FLOPs-efficiency of a 1.42x speed-up.
K-core decomposition is a commonly used metric to analyze graph structure or study the relative importance of nodes in complex graphs. Recent years have seen rapid growth in the scale of the graph, especially in industrial settings. For example, our industrial partner runs popular social applications with billions of users and is able to gather a rich set of user data. As a result, applying K-core decomposition on large graphs has attracted more and more attention from academics and the industry. A simple but effective method to deal with large graphs is to train them in the distributed settings, and some distributed K-core decomposition algorithms are also proposed. Despite their effectiveness, we experimentally and theoretically observe that these algorithms consume too many resources and become unstable on super-large-scale graphs, especially when the given resources are limited. In this paper, we deal with those super-large-scale graphs and propose a divide-and-conquer strategy on top of the distributed K-core decomposition algorithm. We evaluate our approach on three large graphs. The experimental results show that the consumption of resources can be significantly reduced, and the calculation on large-scale graphs becomes more stable than the existing methods. For example, the distributed K-core decomposition algorithm can scale to a large graph with 136 billion edges without losing correctness with our divide-and-conquer technique.
The ensemble of deep neural networks has been shown, both theoretically and empirically, to improve generalization accuracy on the unseen test set. However, the high training cost hinders its efficiency since we need a sufficient number of base models and each one in the ensemble has to be separately trained. Lots of methods are proposed to tackle this problem, and most of them are based on the feature that a pre-trained network can transfer its knowledge to the next base model and then accelerate the training process. However, these methods suffer a severe problem that all of them transfer knowledge without selection and thus lead to low diversity. As the effect of ensemble learning is more pronounced if ensemble members are accurate and diverse, we propose a method named Efficient Diversity-Driven Ensemble (EDDE) to address both the diversity and the efficiency of an ensemble. To accelerate the training process, we propose a novel knowledge transfer method which can selectively transfer the previous generic knowledge. To enhance diversity, we first propose a new diversity measure, then use it to define a diversity-driven loss function for optimization. At last, we adopt a Boosting-based framework to combine the above operations, such a method can also further improve diversity. We evaluate EDDE on Computer Vision (CV) and Natural Language Processing (NLP) tasks. Compared with other well-known ensemble methods, EDDE can get highest ensemble accuracy with the lowest training cost, which means it is efficient in the ensemble of neural networks.
Embedding models have been an effective learning paradigm for high-dimensional data. However, one open issue of embedding models is that their representations (latent factors) often result in large parameter space. We observe that existing distributed training frameworks face a scalability issue of embedding models since updating and retrieving the shared embedding parameters from servers usually dominates the training cycle. In this paper, we propose HET, a new system framework that significantly improves the scalability of huge embedding model training. We embrace skewed popularity distributions of embeddings as a performance opportunity and leverage it to address the communication bottleneck with an embedding cache. To ensure consistency across the caches, we incorporate a new consistency model into HET design, which provides fine-grained consistency guarantees on a per-embedding basis. Compared to previous work that only allows staleness for read operations, HET also utilizes staleness for write operations. Evaluations on six representative tasks show that HET achieves up to 88% embedding communication reductions and up to 20.68x performance speedup over the state-of-the-art baselines.
Recently, zero-shot and few-shot learning via Contrastive Vision-Language Pre-training (CLIP) have shown inspirational performance on 2D visual recognition, which learns to match images with their corresponding texts in open-vocabulary settings. However, it remains under explored that whether CLIP, pre-trained by large-scale image-text pairs in 2D, can be generalized to 3D recognition. In this paper, we identify such a setting is feasible by proposing PointCLIP, which conducts alignment between CLIP-encoded point cloud and 3D category texts. Specifically, we encode a point cloud by projecting it into multi-view depth maps without rendering, and aggregate the view-wise zero-shot prediction to achieve knowledge transfer from 2D to 3D. On top of that, we design an inter-view adapter to better extract the global feature and adaptively fuse the few-shot knowledge learned from 3D into CLIP pre-trained in 2D. By just fine-tuning the lightweight adapter in the few-shot settings, the performance of PointCLIP could be largely improved. In addition, we observe the complementary property between PointCLIP and classical 3D-supervised networks. By simple ensembling, PointCLIP boosts baseline's performance and even surpasses state-of-the-art models. Therefore, PointCLIP is a promising alternative for effective 3D point cloud understanding via CLIP under low resource cost and data regime. We conduct thorough experiments on widely-adopted ModelNet10, ModelNet40 and the challenging ScanObjectNN to demonstrate the effectiveness of PointCLIP. The code is released at https://github.com/ZrrSkywalker/PointCLIP.
Message passing is the core of most graph models such as Graph Convolutional Network (GCN) and Label Propagation (LP), which usually require a large number of clean labeled data to smooth out the neighborhood over the graph. However, the labeling process can be tedious, costly, and error-prone in practice. In this paper, we propose to unify active learning (AL) and message passing towards minimizing labeling costs, e.g., making use of few and unreliable labels that can be obtained cheaply. We make two contributions towards that end. First, we open up a perspective by drawing a connection between AL enforcing message passing and social influence maximization, ensuring that the selected samples effectively improve the model performance. Second, we propose an extension to the influence model that incorporates an explicit quality factor to model label noise. In this way, we derive a fundamentally new AL selection criterion for GCN and LP--reliable influence maximization (RIM)--by considering quantity and quality of influence simultaneously. Empirical studies on public datasets show that RIM significantly outperforms current AL methods in terms of accuracy and efficiency.