Large language models (LLMs) have made significant strides in code translation tasks. However, ensuring both the correctness and readability of translated code remains a challenge, limiting their effective adoption in real-world software development. In this work, we propose F2STrans, a function-to-style guiding paradigm designed to progressively improve the performance of LLMs in code translation. Our approach comprises two key stages: (1) Functional learning, which optimizes translation correctness using high-quality source-target code pairs mined from online programming platforms, and (2) Style learning, which improves translation readability by incorporating both positive and negative style examples. Additionally, we introduce a novel code translation benchmark that includes up-to-date source code, extensive test cases, and manually annotated ground-truth translations, enabling comprehensive functional and stylistic evaluations. Experiments on both our new benchmark and existing datasets demonstrate that our approach significantly improves code translation performance. Notably, our approach enables Qwen-1.5B to outperform prompt-enhanced Qwen-32B and GPT-4 on average across 20 diverse code translation scenarios.

Graph anomaly detection (GAD) aims to find network elements (e.g., nodes, edges) with significantly atypical patterns and has a profound impact in a variety of application domains, including social network analysis, security, Web, finance, and many more. Most of the existing methods have been developed in an unsupervised manner or with extremely limited supervision, due to the high cost of acquiring ground-truth information. Consequently, the identified anomalies may turn out to be noises or uneventful instances because of the lack of prior knowledge on graph anomalies. To address the data scarcity issue in GAD, in this paper, we propose, gADAM, a novel graph neural network-based GAD framework, which consolidates (1) an innovative mixup approach to augment the original training data by adaptively interpolating data instances in the embedding space, and (2) an efficacious sampling method to obtain high-quality negative samples for model training. Additionally, to advance the representation learning for GAD, we further equip the proposed framework with a generic prototype-based learning module. Through extensive empirical evaluations, we corroborate the superiority of the proposed gADAM framework on graph anomaly detection w.r.t. various metrics.

Imbalanced datasets are commonly observed in various real-world applications, presenting significant challenges in training classifiers. When working with large datasets, the imbalanced issue can be further exacerbated, making it exceptionally difficult to train classifiers effectively. To address the problem, over-sampling techniques have been developed to linearly interpolating data instances between minorities and their neighbors. However, in many real-world scenarios such as anomaly detection, minority instances are often dispersed diversely in the feature space rather than clustered together. Inspired by domain-agnostic data mix-up, we propose generating synthetic samples iteratively by mixing data samples from both minority and majority classes. It is non-trivial to develop such a framework, the challenges include source sample selection, mix-up strategy selection, and the coordination between the underlying model and mix-up strategies. To tackle these challenges, we formulate the problem of iterative data mix-up as a Markov decision process (MDP) that maps data attributes onto an augmentation strategy. To solve the MDP, we employ an actor-critic framework to adapt the discrete-continuous decision space. This framework is utilized to train a data augmentation policy and design a reward signal that explores classifier uncertainty and encourages performance improvement, irrespective of the classifier's convergence. We demonstrate the effectiveness of our proposed framework through extensive experiments conducted on seven publicly available benchmark datasets using three different types of classifiers. The results of these experiments showcase the potential and promise of our framework in addressing imbalanced datasets with diverse minorities.

Transformer-based models are powerful for modeling temporal dynamics of user preference in sequential recommendation. Most of the variants adopt the Softmax transformation in the self-attention layers to generate dense attention probabilities. However, real-world item sequences are often noisy, containing a mixture of true-positive and false-positive interactions. Such dense attentions inevitably assign probability mass to noisy or irrelevant items, leading to sub-optimal performance and poor explainability. Here we propose a Probabilistic Masked Attention Network (PMAN) to identify the sparse pattern of attentions, which is more desirable for pruning noisy items in sequential recommendation. Specifically, we employ a probabilistic mask to achieve sparse attentions under a constrained optimization framework. As such, PMAN allows to select which information is critical to be retained or dropped in a data-driven fashion. Experimental studies on real-world benchmark datasets show that PMAN is able to improve the performance of Transformers significantly.

The huge volume of emerging graph datasets has become a double-bladed sword for graph machine learning. On the one hand, it empowers the success of a myriad of graph neural networks (GNNs) with strong empirical performance. On the other hand, training modern graph neural networks on huge graph data is computationally expensive. How to distill the given graph dataset while retaining most of the trained models' performance is a challenging problem. Existing efforts try to approach this problem by solving meta-learning-based bilevel optimization objectives. A major hurdle lies in that the exact solutions of these methods are computationally intensive and thus, most, if not all, of them are solved by approximate strategies which in turn hurt the distillation performance. In this paper, inspired by the recent advances in neural network kernel methods, we adopt a kernel ridge regression-based meta-learning objective which has a feasible exact solution. However, the computation of graph neural tangent kernel is very expensive, especially in the context of dataset distillation. As a response, we design a graph kernel, named LiteGNTK, tailored for the dataset distillation problem which is closely related to the classic random walk graph kernel. An effective model named Kernel rıdge regression-based graph Dataset Distillation (KIDD) and its variants are proposed. KIDD shows nice efficiency in both the forward and backward propagation processes. At the same time, KIDD shows strong empirical performance over 7 real-world datasets compared with the state-of-the-art distillation methods. Thanks to the ability to find the exact solution of the distillation objective, the learned training graphs by KIDD can sometimes even outperform the original whole training set with as few as 1.65% training graphs.

Graph neural networks (GNNs) have become core building blocks behind a myriad of graph learning tasks. The vast majority of the existing GNNs are built upon, either implicitly or explicitly, the homophily assumption, which is not always true and could heavily degrade the performance of learning tasks. In response, GNNs tailored for heterophilic graphs have been developed. However, most of the existing works are designed for the specific GNN models to address heterophily, which lacks generality. In this paper, we study the problem from the structure learning perspective and propose a family of general solutions named ALT. It can work hand in hand with most of the existing GNNs to handle graphs with either low or high homophily. At the core of our method is learning to (1) decompose a given graph into two components, (2) extract complementary graph signals from these two components, and (3) adaptively integrate the graph signals for node classification. Moreover, analysis based on graph signal processing shows that our framework can empower a broad range of existing GNNs to have adaptive filter characteristics and further modulate the input graph signals, which is critical for handling complex homophilic/heterophilic patterns. The proposed ALT brings significant and consistent performance improvement in node classification for a wide range of GNNs over a variety of real-world datasets.

Finding the proper depth $d$ of a graph convolutional network (GCN) that provides strong representation ability has drawn significant attention, yet nonetheless largely remains an open problem for the graph learning community. Although noteworthy progress has been made, the depth or the number of layers of a corresponding GCN is realized by a series of graph convolution operations, which naturally makes $d$ a positive integer ($d \in \mathbb{N}+$). An interesting question is whether breaking the constraint of $\mathbb{N}+$ by making $d$ a real number ($d \in \mathbb{R}$) can bring new insights into graph learning mechanisms. In this work, by redefining GCN's depth $d$ as a trainable parameter continuously adjustable within $(-\infty,+\infty)$, we open a new door of controlling its signal processing capability to model graph homophily/heterophily (nodes with similar/dissimilar labels/attributes tend to be inter-connected). A simple and powerful GCN model TEDGCN, is proposed to retain the simplicity of GCN and meanwhile automatically search for the optimal $d$ without the prior knowledge regarding whether the input graph is homophilic or heterophilic. Negative-valued $d$ intrinsically enables high-pass frequency filtering functionality via augmented topology for graph heterophily. Extensive experiments demonstrate the superiority of TEDGCN on node classification tasks for a variety of homophilic and heterophilic graphs.

Collaborative Filtering (CF) has been successfully used to help users discover the items of interest. Nevertheless, existing CF methods suffer from noisy data issue, which negatively impacts the quality of recommendation. To tackle this problem, many prior studies leverage adversarial learning to regularize the representations of users/items, which improves both generalizability and robustness. Those methods often learn adversarial perturbations and model parameters under min-max optimization framework. However, there still have two major drawbacks: 1) Existing methods lack theoretical guarantees of why adding perturbations improve the model generalizability and robustness; 2) Solving min-max optimization is time-consuming. In addition to updating the model parameters, each iteration requires additional computations to update the perturbations, making them not scalable for industry-scale datasets.

Graph Neural Networks (GNNs) have achieved state-of-the-art performance in recommender systems. Nevertheless, the process of searching and ranking from a large item corpus usually requires high latency, which limits the widespread deployment of GNNs in industry-scale applications. To address this issue, many methods compress user/item representations into the binary embedding space to reduce space requirements and accelerate inference. Also, they use the Straight-through Estimator (STE) to prevent vanishing gradients during back-propagation. However, the STE often causes the gradient mismatch problem, leading to sub-optimal results.

Transformer and its variants are a powerful class of architectures for sequential recommendation, owing to their ability of capturing a user’s dynamic interests from their past interactions. Despite their success, Transformer-based models often require the optimization of a large number of parameters, making them difficult to train from sparse data in sequential recommendation. To address the problem of data sparsity, previous studies have utilized self-supervised learning to enhance Transformers, such as pre-training embeddings from item attributes or contrastive data augmentations. However, these approaches encounter several training issues, including initialization sensitivity, manual data augmentations, and large batch-size memory bottlenecks.

Time series anomaly detection is a challenging task with a wide range of real-world applications. Due to label sparsity, training a deep anomaly detector often relies on unsupervised approaches. Recent efforts have been devoted to time series domain adaptation to leverage knowledge from similar domains. However, existing solutions may suffer from negative knowledge transfer on anomalies due to their diversity and sparsity. Motivated by the empirical study of context alignment between two domains, we aim to transfer knowledge between two domains via adaptively sampling context information for two domains. This is challenging because it requires simultaneously modeling the complex in-domain temporal dependencies and cross-domain correlations while exploiting label information from the source domain. To this end, we propose a framework that combines context sampling and anomaly detection into a joint learning procedure. We formulate context sampling into the Markov decision process and exploit deep reinforcement learning to optimize the time series domain adaptation process via context sampling and design a tailored reward function to generate domain-invariant features that better align two domains for anomaly detection. Experiments on three public datasets show promise for knowledge transfer between two similar domains and two entirely different domains.

Graph Neural Networks (GNNs) have achieved impressive performance in collaborative filtering. However, recent studies show that GNNs tend to yield inferior performance when the distributions of training and test data are not aligned well. Moreover, training GNNs often requires optimizing non-convex neural networks with an abundance of local and global minima, which may differ widely in their performance at test time. Thus, it is essential to develop an optimization strategy that can choose the minima carefully, which can yield strong generalization performance on unseen data. Here we propose an effective training schema, called gSAM, under the principle that theflatter minima has a better generalization ability than thesharper ones. To achieve this goal, gSAM regularizes the flatness of the weight loss landscape by forming a bi-level optimization: the outer problem conducts the standard model training while the inner problem helps the model jump out of the sharp minima. Experimental results show the superiority of our gSAM.

Knowledge graph question answering aims to identify answers of the query according to the facts in the knowledge graph. In the vast majority of the existing works, the input queries are considered perfect and can precisely express the user’s query intention. However, in reality, input queries might be ambiguous and elusive which only contain a limited amount of information. Directly answering these ambiguous queries may yield unwanted answers and deteriorate user experience. In this paper, we propose PReFNet which focuses on answering ambiguous queries with pseudo relevance feedback on knowledge graphs. In order to leverage the hidden (pseudo) relevance information existed in the results that are initially returned from a given query, PReFNet treats the top-k returned candidate answers as a set of most relevant answers, and uses variational Bayesian inference to infer user’s query intention. To boost the quality of the inferred queries, a neighborhood embedding based VGAE model is used to prune inferior inferred queries. The inferred high quality queries will be returned to the users to help them search with ease. Moreover, all the high-quality candidate nodes will be re-ranked according to the inferred queries. The experiment results show that our proposed method can recommend high-quality query graphs to users and improve the question answering accuracy.

Node embeddings obtained from information networks have been widely adopted for representing knowledge and driving various information retrieval and machine learning tasks. However, training node embeddings is computationally intensive, making it difficult to scale to larger graphs. Most existing works have addressed the scalability challenge by simply adding more hardware resources. For example, a common approach to speed up the training process is to distribute model computation across multiple machines and GPUs. This paper takes an orthogonal approach towards scalability by addressing the problem of computation complexity in training embeddings. We present bigg2vec for scaling up the embedding training process. bigg2vec introduces a novel polar coordinate-based system for internal representation and computation. It provides the following benefits: (a) It significantly reduces compute and memory requirements while improving embedding quality and (b) uses a novel graph organization to generate high-quality negative samples (this reduces the number of negative samples needed for training, which is especially beneficial f or skewed graphs). We have deployed bigg2vec to generate embeddings for multiple AI models within Visa. Our Global Personalized Restaurant Recommender System (GPR) is one such project that uses bigg2vec to periodically generate embeddings for over 450 million nodes connected by more than 3 billion edges. bigg2vec generates higher quality embeddings while training them faster than state-of-the-art methods on a single CPU-based machine.

Graph neural networks have achieved significant success in representation learning. However, the performance gains come at a cost; acquiring comprehensive labeled data for training can be prohibitively expensive. Active learning mitigates this issue by searching the unexplored data space and prioritizing the selection of data to maximize model's performance gain. In this paper, we propose a novel method SMARTQUERY, a framework to learn a graph neural network with very few labeled nodes using a hybrid uncertainty reduction function. This is achieved using two key steps: (a) design a multi-stage active graph learning framework by exploiting diverse explicit graph information and (b) introduce label propagation to efficiently exploit known labels to assess the implicit embedding information. Using a comprehensive set of experiments on three network datasets, we demonstrate the competitive performance of our method against state-of-the-arts on very few labeled data (up to 5 labeled nodes per class).

Real-world link prediction problems often deal with data from multiple domains, where data may be highly skewed and imbalanced. Computer vision research has studied similar issues under the Few-Shot Learning (FSL) umbrella. However, this problem has rarely been addressed and explored in the graph domain, specifically for link prediction. In this work, we propose an adversarial training-based framework that aims at improving link prediction for highly skewed and imbalanced graphs from different domains by generating domain agnostic embedding. We introduce a domain discriminator on pairs of graph-level embedding. We then use the discriminator to improve the model in an adversarial way, such that the graph embedding generated by the model are domain agnostic. We test our ideas on one large real-world user-business-review dataset and three benchmark datasets. Our results demonstrate that when domain differences exist, our method creates better graph embedding that are more evenly distributed across domains and generate better prediction outcomes. In the absence of domain differences, our method is on par with state-of-the-art.

Transformer-based sequential recommenders are very powerful for capturing both short-term and long-term sequential item dependencies. This is mainly attributed to their unique self-attention networks to exploit pairwise item-item interactions within the sequence. However, real-world item sequences are often noisy, which is particularly true for implicit feedback. For example, a large portion of clicks do not align well with user preferences, and many products end up with negative reviews or being returned. As such, the current user action only depends on a subset of items, not on the entire sequences. Many existing Transformer-based models use full attention distributions, which inevitably assign certain credits to irrelevant items. This may lead to sub-optimal performance if Transformers are not regularized properly.

There has been an explosion of interest in designing various Knowledge Graph Neural Networks (KGNNs), which achieve state-of-the-art performance and provide great explainability for recommendation. The promising performance is mainly resulting from their capability of capturing high-order proximity messages over the knowledge graphs. However, training KGNNs at scale is challenging due to the high memory usage. In the forward pass, the automatic differentiation engines (e.g., TensorFlow/PyTorch) generally need to cache all intermediate activation maps in order to compute gradients in the backward pass, which leads to a large GPU memory footprint. Existing work solves this problem by utilizing multi-GPU distributed frameworks. Nonetheless, this poses a practical challenge when seeking to deploy KGNNs in memory-constrained environments, especially for industry-scale graphs.

Graph Neural Networks (GNNs) provide a class of powerful architectures that are effective for graph-based collaborative filtering. Nevertheless, GNNs are known to be vulnerable to adversarial perturbations. Adversarial training is a simple yet effective way to improve the robustness of neural models. For example, many prior studies inject adversarial perturbations into either node features or hidden layers of GNNs. However, perturbing graph structures has been far less studied in recommendations.

Graph Neural Networks (GNNs) have emerged as powerful tools for collaborative filtering. A key challenge of recommendations is to distill long-range collaborative signals from user-item graphs. Typically, GNNs generate embeddings of users/items by propagating and aggregating the messages between local neighbors. Thus, the ability of GNNs to capture long-range dependencies heavily depends on their depths. However, simply training deep GNNs has several bottleneck effects, e.g., over-fitting & over-smoothing, which may lead to unexpected results if GNNs are not well regularized.

Deep reinforcement learning (DRL) targets to train an autonomous agent to interact with a pre-defined environment and strives to achieve specific goals through deep neural networks (DNN). Recurrent neural network (RNN) based DRL has demonstrated superior performance, as RNNs can effectively capture the temporal evolution of the environment and respond with proper agent actions. However, apart from the outstanding performance, little is known about how RNNs understand the environment internally and what has been memorized over time. Revealing these details is extremely important for deep learning experts to understand and improve DRLs, which in contrast, is also challenging due to the complicated data transformations inside these models. In this article, we propose Deep Reinforcement Learning Interactive Visual Explorer (DRLIVE), a visual analytics system to effectively explore, interpret, and diagnose RNN-based DRLs. Having focused on DRL agents trained for different Atari games, DRLIVE accomplishes three tasks: game episode exploration, RNN hidden/cell state examination, and interactive model perturbation. Using the system, one can flexibly explore a DRL agent through interactive visualizations, discover interpretable RNN cells by prioritizing RNN hidden/cell states with a set of metrics, and further diagnose the DRL model by interactively perturbing its inputs. Through concrete studies with multiple deep learning experts, we validated the efficacy of DRLIVE.

Adversarial examples are input examples that are specifically crafted to deceive machine learning classifiers. State-of-the-art adversarial example detection methods characterize an input example as adversarial either by quantifying the magnitude of feature variations under multiple perturbations or by measuring its distance from estimated benign example distribution. Instead of using such metrics, the proposed method is based on the observation that the directions of adversarial gradients when crafting (new) adversarial examples play a key role in characterizing the adversarial space. Compared to detection methods that use multiple perturbations, the proposed method is efficient as it only applies a single random perturbation on the input example. Experiments conducted on two different databases, CIFAR-10 and ImageNet, show that the proposed detection method achieves, respectively, 97.9% and 98.6% AUC-ROC (on average) on five different adversarial attacks, and outperforms multiple state-of-the-art detection methods. Results demonstrate the effectiveness of using adversarial gradient directions for adversarial example detection.

Tree boosting models are widely adopted predictive models and have demonstrated superior performance than other conventional and even deep learning models, especially since the recent release of their parallel and distributed implementations, e.g., XGBoost, LightGMB, and CatBoost. Tree boosting uses a group of sequentially generated weak learners (i.e., decision trees), each learns from the mistakes of its predecessor, to push the model’s decision boundary towards the true boundary. As the number of trees keeps increasing over training, it is important to reveal how the newly-added trees change the predictions of individual data instances, and how the impacts of different data features evolve. To accomplish these goals, in this paper, we introduce a new design of the temporal confusion matrix, providing users with an effective interface to track data instances’ predictions across the tree boosting process. Also, we present an improved visualization to better illustrate and compare the impacts of individual data features (based on their SHAP values) across training iterations. Integrating these components with a tree structure visualization component, we propose a visual analytics system for tree boosting models. Through case studies with domain experts using real-world datasets, we validated the system’s effectiveness.

Today’s cyber-world is vastly multivariate. Metrics collected at extreme varieties demand multivariate algorithms to properly detect anomalies. However, forecast-based algorithms, as widely proven approaches, often perform sub-optimally or inconsistently across datasets. A key common issue is they strive to be one-size-fits-all but anomalies are distinctive in nature. We propose a method that tailors to such distinction. Presenting FMUAD - a Forecast-based, Multi-aspect, Unsupervised Anomaly Detection framework. FMUAD explicitly and separately captures the signature traits of anomaly types - spatial change, temporal change and correlation change - with independent modules. The modules then jointly learn an optimal feature representation, which is highly flexible and intuitive, unlike most other models in the category. Extensive experiments show our FMUAD framework consistently outperforms other state-of-the-art forecast-based anomaly detectors.

Detection of adversarial examples with high accuracy is critical for the security of deployed deep neural network-based models. We present the first graph-based adversarial detection method that constructs a Latent Neighborhood Graph (LNG) around an input example to determine if the input example is adversarial. Given an input example, selected reference adversarial and benign examples (represented as LNG nodes in Figure 1) are used to capture the local manifold in the vicinity of the input example. The LNG node connectivity parameters are optimized jointly with the parameters of a graph attention network in an end-to-end manner to determine the optimal graph topology for adversarial example detection. The graph attention network is used to determine if the LNG is derived from an adversarial or benign input example. Experimental evaluations on CIFAR-10, STL-10, and ImageNet datasets, using six adversarial attack methods, demonstrate that the proposed method outperforms state-of-the-art adversarial detection methods in white-box and gray-box settings. The proposed method is able to successfully detect adversarial examples crafted with small perturbations using unseen attacks.

Fashion is more than Paris runways. Fashion is about how people express their interests, identity, mood, and cultural influences. Given an inventory of candidate garments from different categories, how to assemble them together would most improve their fashionability? This question presents an intriguing visual recommendation challenge to automatically create capsule wardrobes. Capsule wardrobe generation is a complex combinatorial problem that requires the understanding of how multiple visual items interact. The generative process often needs fashion experts to manually tease the combinations out, making it hard to scale.

Graph Convolutional Networks (GCNs) are powerful for collaborative filtering. The key component of GCNs is to explore neighborhood aggregation mechanisms to extract high-level representations of users and items. However, real-world user-item graphs are often incomplete and noisy. Aggregating misleading neighborhood information may lead to sub-optimal performance if GCNs are not regularized properly. Also, the real-world user-item graphs are often sparse and low rank. These two intrinsic graph properties are widely used in shallow matrix completion models, but far less studied in graph neural models. Here we propose Structured Graph Convolutional Networks (SGCNs) to enhance the performance of GCNs by exploiting graph structural properties of sparsity and low rank. To achieve sparsity, we attach each layer of a GCN with a trainable stochastic binary mask to prune noisy and insignificant edges, resulting in a clean and sparsified graph. To preserve its low-rank property, the nuclear norm regularization is applied. We jointly learn the parameters of stochastic binary masks and original GCNs by solving a stochastic binary optimization problem. An unbiased gradient estimator is further proposed to better backpropagate the gradients of binary variables. Experimental results demonstrate that SGCNs achieve better performance compared with the state-of-the-art GCNs.

In this paper, we demonstrate our Global Personalized Recommender (GPR) system for restaurants. GPR does not use any explicit reviews, ratings, or domain-specific metadata but rather leverages over 3 billion anonymized payment transactions to learn user and restaurant behavior patterns. The design and development of GPR have been challenging, primarily due to the scale and skew of the data. Our system supports over 450M cardholders from over 200 countries and 2.5M restaurants in over 35K cities worldwide, respectively. Additionally, GPR being a global recommender system, needs to account for the regional variations in people's food choices and habits. We address the challenges by combining three different recommendation algorithms instead of using a single revolutionary model in the backend. The individual recommendation models are scalable and adapt to varying data skew challenges to ensure high-quality personalized recommendations for any user anywhere in the world.

Two fundamental problems in machine learning are recognition and generation. Apart from the tremendous amount of research efforts devoted to these two problems individually, finding the association between them has attracted increasingly more attention recently. Symbol-Concept Association Network (SCAN) is one of the most popular models for this problem proposed by Google DeepMind lately, which integrates an unsupervised concept abstraction process and a supervised symbol-concept association process. Despite the outstanding performance of this deep neural network, interpreting and evaluating it remain challenging. Guided by the practical needs from deep learning experts, this paper proposes a visual analytics attempt, i.e., SCANViz, to address this challenge in the visual domain. Specifically, SCANViz evaluates the performance of SCAN through its power of recognition and generation, facilitates the exploration of the latent space derived from both the unsupervised extraction and supervised association processes, empowers interactive training of SCAN to interpret the model’s understanding on a particular visual concept. Through concrete case studies with multiple deep learning experts, we validate the effectiveness of SCANViz.

We present Magellan - a personalized travel recommendation system that is built entirely from card transaction data. The data logs contain extensive metadata for each transaction between a user and a merchant. We describe the procedure employed to extract travel itineraries from such transaction data. Unlike traditional approaches, we formulate the recommendation problem into two steps: (1) predict coarse granularity information such as location and category of the next merchant; and (2) provide fine granularity individual merchant recommendations based on the predicted location and category. The breakdown helps us build a scalable recommendation system. We propose a quadtree-based algorithm that provides an adaptive spatial resolution for the location classes in our first step while also reducing the class-imbalance across various location labels. Finally, we propose a novel neural architecture, SoLEmNet, that implicitly learns the inherent class label hierarchy and achieves a higher performance on our dataset compared to previous baselines.

The rapid proliferation of new users and items on the social web has aggravated the gray-sheep user/long-tail item challenge in recommender systems. Historically, cross-domain co-clustering methods have successfully leveraged shared users and items across dense and sparse domains to improve inference quality. However, they rely on shared rating data and cannot scale to multiple sparse target domains (i.e., the one-to-many transfer setting). This, combined with the increasing adoption of neural recommender architectures, motivates us to develop scalable neural layer-transfer approaches for cross-domain learning. Our key intuition is to guide neural collaborative filtering with domain-invariant components shared across the dense and sparse domains, improving the user and item representations learned in the sparse domains. We leverage contextual invariances across domains to develop these shared modules, and demonstrate that with user-item interaction context, we can learn-to-learn informative representation spaces even with sparse interaction data. We show the effectiveness and scalability of our approach on two public datasets and a massive transaction dataset from Visa, a global payments technology company (19% Item Recall, 3x faster vs. training separate models for each domain). Our approach is applicable to both implicit and explicit feedback settings.

We study the problem of making item recommendations to ephemeral groups, which comprise users with limited or no historical activities together. Existing studies target persistent groups with substantial activity history, while ephemeral groups lack historical interactions. To overcome group interaction sparsity, we propose data-driven regularization strategies to exploit both the preference covariance amongst users who are in the same group, as well as the contextual relevance of users' individual preferences to each group.

Learning node representations in graphs is important for many applications such as link prediction, node classification, and community detection. Existing graph representation learning methods primarily target static graphs while many real-world graphs evolve over time. Complex time-varying graph structures make it challenging to learn informative node representations over time.

Recommending fashion outfits to users presents several challenges. First of all, an outfit consists of multiple fashion items, and each user emphasizes different parts of an outfit when considering whether they like it or not. Secondly, a user’s liking for a fashion outfit considers not only the aesthetics of each item but also the compatibility among them. Lastly, fashion outfit data is often sparse in terms of the relationship between users and fashion outfits. Not to mention, we can only obtain what the users like, but not what they dislike.

Recent years have witnessed a surge of interest in machine learning on graphs and networks with applications ranging from IoT traffic management to social network recommendations. Supervised machine learning tasks in networks such as node classification and link prediction require us to perform feature engineering that is known and agreed to be the key to success in applied machine learning. Research efforts dedicated to representation learning, especially representation learning using deep learning, has shown us ways to automatically learn relevant features from vast amounts of potentially noisy, raw data. However, most of the methods are inadequate to handle heterogeneous information networks which pretty much represents most real world data today. The methods cannot preserve the structure and semantic of multiple types of nodes and links well enough, capture higher-order heterogeneous connectivity patterns, and ensure coverage of nodes for which representations are generated. In this paper, we propose a novel efficient algorithm, motif2vec that learns node representations or embeddings for heterogeneous networks. Specifically, we leverage higher-order, recurring, and statistically significant network connectivity patterns in the form of motifs to transform the original graph to motif graph(s), conduct biased random walk to efficiently explore higher order neighborhoods, and then employ heterogeneous skip-gram model to generate the embeddings. We evaluate the proposed algorithm on multiple real-world networks from diverse domains and against existing state-of-the-art methods on multi-class node classification and link prediction tasks, and demonstrate its consistent superiority over prior work.

Deep Q-Network (DQN), as one type of deep reinforcement learning model, targets to train an intelligent agent that acquires optimal actions while interacting with an environment. The model is well known for its ability to surpass professional human players across many Atari 2600 games. Despite the superhuman performance, in-depth understanding of the model and interpreting the sophisticated behaviors of the DQN agent remain to be challenging tasks, due to the long-time model training process and the large number of experiences dynamically generated by the agent. In this work, we propose DQNViz, a visual analytics system to expose details of the blind training process in four levels, and enable users to dive into the large experience space of the agent for comprehensive analysis. As an initial attempt in visualizing DQN models, our work focuses more on Atari games with a simple action space, most notably the Breakout game. From our visual analytics of the agent's experiences, we extract useful action/reward patterns that help to interpret the model and control the training. Through multiple case studies conducted together with deep learning experts, we demonstrate that DQNViz can effectively help domain experts to understand, diagnose, and potentially improve DQN models.

Deep Neural Networks (DNNs) have been extensively used in multiple disciplines due to their superior performance. However, in most cases, DNNs are considered as black-boxes and the interpretation of their internal working mechanism is usually challenging. Given that model trust is often built on the understanding of how a model works, the interpretation of DNNs becomes more important, especially in safety-critical applications (e.g., medical diagnosis, autonomous driving). In this paper, we propose DeepVID, a Deep learning approach to Visually Interpret and Diagnose DNN models, especially image classifiers. In detail, we train a small locally-faithful model to mimic the behavior of an original cumbersome DNN around a particular data instance of interest, and the local model is sufficiently simple such that it can be visually interpreted (e.g., a linear model). Knowledge distillation is used to transfer the knowledge from the cumbersome DNN to the small model, and a deep generative model (i.e., variational auto-encoder) is used to generate neighbors around the instance of interest. Those neighbors, which come with small feature variances and semantic meanings, can effectively probe the DNN's behaviors around the interested instance and help the small model to learn those behaviors. Through comprehensive evaluations, as well as case studies conducted together with deep learning experts, we validate the effectiveness of DeepVID.

Generative models bear promising implications to learn data representations in an unsupervised fashion with deep learning. Generative Adversarial Nets (GAN) is one of the most popular frameworks in this arena. Despite the promising results from different types of GANs, in-depth understanding on the adversarial training process of the models remains a challenge to domain experts. The complexity and the potential long-time training process of the models make it hard to evaluate, interpret, and optimize them. In this work, guided by practical needs from domain experts, we design and develop a visual analytics system, GANViz, aiming to help experts understand the adversarial process of GANs in-depth. Specifically, GANViz evaluates the model performance of two subnetworks of GANs, provides evidence and interpretations of the models' performance, and empowers comparative analysis with the evidence. Through our case studies with two real-world datasets, we demonstrate that GANViz can provide useful insight into helping domain experts understand, interpret, evaluate, and potentially improve GAN models.

Recently, many e-commerce websites have encouraged their users to rate shopping items and write review texts. This review information has been very useful for understanding user preferences and item properties, as well as enhancing the capability to make personalized recommendations of these websites. In this paper, we propose to model user preferences and item properties using convolutional neural networks (CNNs) with dual local and global attention, motivated by the superiority of CNNs to extract complex features. By using aggregated review texts from a user and aggregated review text for an item, our model can learn the unique features (embedding) of each user and each item. These features are then used to predict ratings. We train these user and item networks jointly which enable the interaction between users and items in a similar way as matrix factorization. The local attention provides us insight on a user's preferences or an item's properties. The global attention helps CNNs focus on the semantic meaning of the whole review text. Thus, the combined local and global attentions enable an interpretable and better-learned representation of users and items. We validate the proposed models by testing on popular review datasets in Yelp and Amazon and compare the results with matrix factorization (MF), the hidden factor and topical (HFT) model, and the recently proposed convolutional matrix factorization (ConvMF+). Our proposed CNNs with dual attention model outperforms HFT and ConvMF+ in terms of mean square errors (MSE). In addition, we compare the user/item embeddings learned from these models for classification and recommendation. These results also confirm the superior quality of user/item embeddings learned from our model.