Learning the distance metric between pairs of examples is of great importance for learning and visual recognition. With the remarkable success from the state of the art convolutional neural networks, recent works [1, 31] have shown promising results on discriminatively training the networks to learn semantic feature embeddings where similar examples are mapped close to each other and dissimilar examples are mapped farther apart. In this paper, we describe an algorithm for taking full advantage of the training batches in the neural network training by lifting the vector of pairwise distances within the batch to the matrix of pairwise distances. This step enables the algorithm to learn the state of the art feature embedding by optimizing a novel structured prediction objective on the lifted problem. Additionally, we collected Stanford Online Products dataset: 120k images of 23k classes of online products for metric learning. Our experiments on the CUB-200-2011 [37], CARS196 [19], and Stanford Online Products datasets demonstrate significant improvement over existing deep feature embedding methods on all experimented embedding sizes with the GoogLeNet [33] network. The source code and the dataset are available at: https://github.com/rksltnl/ Deep-Metric-Learning-CVPR16.

/pdf/deep-metric-learning-via-lifted-structured-feature-embedding-2537kz8vq3.pdf

Deep Metric Learning via Lifted Structured Feature Embedding

Unsupervised text embedding methods, such as Skip-gram and Paragraph Vector, have been attracting increasing attention due to their simplicity, scalability, and effectiveness. However, comparing to sophisticated deep learning architectures such as convolutional neural networks, these methods usually yield inferior results when applied to particular machine learning tasks. One possible reason is that these text embedding methods learn the representation of text in a fully unsupervised way, without leveraging the labeled information available for the task. Although the low dimensional representations learned are applicable to many different tasks, they are not particularly tuned for any task. In this paper, we fill this gap by proposing a semi-supervised representation learning method for text data, which we call the predictive text embedding (PTE). Predictive text embedding utilizes both labeled and unlabeled data to learn the embedding of text. The labeled information and different levels of word co-occurrence information are first represented as a large-scale heterogeneous text network, which is then embedded into a low dimensional space through a principled and efficient algorithm. This low dimensional embedding not only preserves the semantic closeness of words and documents, but also has a strong predictive power for the particular task. Compared to recent supervised approaches based on convolutional neural networks, predictive text embedding is comparable or more effective, much more efficient, and has fewer parameters to tune.

/pdf/pte-predictive-text-embedding-through-large-scale-49tam0d00v.pdf

PTE: Predictive Text Embedding through Large-scale Heterogeneous Text Networks

Extreme multi-label text classification (XMTC) refers to the problem of assigning to each document its most relevant subset of class labels from an extremely large label collection, where the number of labels could reach hundreds of thousands or millions. The huge label space raises research challenges such as data sparsity and scalability. Significant progress has been made in recent years by the development of new machine learning methods, such as tree induction with large-margin partitions of the instance spaces and label-vector embedding in the target space. However, deep learning has not been explored for XMTC, despite its big successes in other related areas. This paper presents the first attempt at applying deep learning to XMTC, with a family of new Convolutional Neural Network (CNN) models which are tailored for multi-label classification in particular. With a comparative evaluation of 7 state-of-the-art methods on 6 benchmark datasets where the number of labels is up to 670,000, we show that the proposed CNN approach successfully scaled to the largest datasets, and consistently produced the best or the second best results on all the datasets. On the Wikipedia dataset with over 2 million documents and 500,000 labels in particular, it outperformed the second best method by 11.7%~15.3% in precision@K and by 11.5%~11.7% in NDCG@K for K = 1,3,5.

https://dl.acm.org/doi/pdf/10.1145/3077136.3080834

Deep Learning for Extreme Multi-label Text Classification

The objective in extreme multi-label learning is to train a classifier that can automatically tag a novel data point with the most relevant subset of labels from an extremely large label set. Embedding based approaches attempt to make training and prediction tractable by assuming that the training label matrix is low-rank and reducing the effective number of labels by projecting the high dimensional label vectors onto a low dimensional linear subspace. Still, leading embedding approaches have been unable to deliver high prediction accuracies, or scale to large problems as the low rank assumption is violated in most real world applications.

In this paper we develop the SLEEC classifier to address both limitations. The main technical contribution in SLEEC is a formulation for learning a small ensemble of local distance preserving embeddings which can accurately predict infrequently occurring (tail) labels. This allows SLEEC to break free of the traditional low-rank assumption and boost classification accuracy by learning embeddings which preserve pairwise distances between only the nearest label vectors.

We conducted extensive experiments on several real-world, as well as benchmark data sets and compared our method against state-of-the-art methods for extreme multi-label classification. Experiments reveal that SLEEC can make significantly more accurate predictions then the state-of-the-art methods including both embedding-based (by as much as 35%) as well as tree-based (by as much as 6%) methods. SLEEC can also scale efficiently to data sets with a million labels which are beyond the pale of leading embedding methods.

/pdf/sparse-local-embeddings-for-extreme-multi-label-gu0rhwq7gq.pdf

Sparse local embeddings for extreme multi-label classification

The objective in extreme multi-label classification is to learn a classifier that can automatically tag a data point with the most relevant subset of labels from a large label set. Extreme multi-label classification is an important research problem since not only does it enable the tackling of applications with many labels but it also allows the reformulation of ranking problems with certain advantages over existing formulations. Our objective, in this paper, is to develop an extreme multi-label classifier that is faster to train and more accurate at prediction than the state-of-the-art Multi-label Random Forest (MLRF) algorithm [2] and the Label Partitioning for Sub-linear Ranking (LPSR) algorithm [35]. MLRF and LPSR learn a hierarchy to deal with the large number of labels but optimize task independent measures, such as the Gini index or clustering error, in order to learn the hierarchy. Our proposed FastXML algorithm achieves significantly higher accuracies by directly optimizing an nDCG based ranking loss function. We also develop an alternating minimization algorithm for efficiently optimizing the proposed formulation. Experiments reveal that FastXML can be trained on problems with more than a million labels on a standard desktop in eight hours using a single core and in an hour using multiple cores.

/pdf/fastxml-a-fast-accurate-and-stable-tree-classifier-for-ogk7f71wzf.pdf

FastXML: a fast, accurate and stable tree-classifier for extreme multi-label learning

The choice of the loss function is critical in extreme multi-label learning where the objective is to annotate each data point with the most relevant subset of labels from an extremely large label set. Unfortunately, existing loss functions, such as the Hamming loss, are unsuitable for learning, model selection, hyperparameter tuning and performance evaluation. This paper addresses the issue by developing propensity scored losses which: (a) prioritize predicting the few relevant labels over the large number of irrelevant ones; (b) do not erroneously treat missing labels as irrelevant but instead provide unbiased estimates of the true loss function even when ground truth labels go missing under arbitrary probabilistic label noise models; and (c) promote the accurate prediction of infrequently occurring, hard to predict, but rewarding tail labels. Another contribution is the development of algorithms which efficiently scale to extremely large datasets with up to 9 million labels, 70 million points and 2 million dimensions and which give significant improvements over the state-of-the-art. This paper's results also apply to tagging, recommendation and ranking which are the motivating applications for extreme multi-label learning. They generalize previous attempts at deriving unbiased losses under the restrictive assumption that labels go missing uniformly at random from the ground truth. Furthermore, they provide a sound theoretical justification for popular label weighting heuristics used to recommend rare items. Finally, they demonstrate that the proposed contributions align with real world applications by achieving superior clickthrough rates on sponsored search advertising in Bing.

Extreme Multi-label Loss Functions for Recommendation, Tagging, Ranking & Other Missing Label Applications

Recommending phrases from web pages for advertisers to bid on against search engine queries is an important research problem with direct commercial impact. Most approaches have found it infeasible to determine the relevance of all possible queries to a given ad landing page and have focussed on making recommendations from a small set of phrases extracted (and expanded) from the page using NLP and ranking based techniques. In this paper, we eschew this paradigm, and demonstrate that it is possible to efficiently predict the relevant subset of queries from a large set of monetizable ones by posing the problem as a multi-label learning task with each query being represented by a separate label. We develop Multi-label Random Forests to tackle problems with millions of labels. Our proposed classifier has prediction costs that are logarithmic in the number of labels and can make predictions in a few milliseconds using 10 Gb of RAM. We demonstrate that it is possible to generate training data for our classifier automatically from click logs without any human annotation or intervention. We train our classifier on tens of millions of labels, features and training points in less than two days on a thousand node cluster. We develop a sparse semi-supervised multi-label learning formulation to deal with training set biases and noisy labels harvested automatically from the click logs. This formulation is used to infer a belief in the state of each label for each training ad and the random forest classifier is extended to train on these beliefs rather than the given labels. Experiments reveal significant gains over ranking and NLP based techniques on a large test set of 5 million ads using multiple metrics.

Multi-label learning with millions of labels: recommending advertiser bid phrases for web pages

This paper develops the Parabel algorithm for extreme multi-label learning where the objective is to learn classifiers that can annotate each data point with the most relevant subset of labels from an extremely large label set. The state-of-the-art 1-vs-All based DiSMEC and PPDSparse algorithms are the most accurate but can take upto months for training and prediction as they learn and apply an independent linear classifier per label. Consequently, they do not scale to large datasets with millions of labels. Parabel addresses both limitations by learning a balanced label hierarchy such that: (a) the 1-vs-All classifiers in the leaf nodes of the label hierarchy can be trained on a small subset of the training set thereby reducing the training time to a few hours on a single core of a standard desktop and (b) novel points can be classified by traversing the learned hierarchy in logarithmic time and applying the 1-vs-All classifiers present in just the leaf thereby reducing the prediction time to a few milliseconds per test point. This allows Parabel to scale to tasks considered infeasible for DiSMEC and PPDSparse such as predicting the subset of 7 million Bing queries that might lead to a click on a given ad-landing page for dynamic search advertising. Experiments on multiple benchmark datasets revealed that Parabel could be almost as accurate as PPDSparse and DiSMEC while being upto 1,000x faster at training and upto 40x-10,000x faster at prediction. Furthermore, Parabel was demonstrated to significantly improve dynamic search advertising on Bing by more than doubling the ad recall and improving the click-through rate by 20%. Source code for Parabel can be downloaded from [1].

Parabel: Partitioned Label Trees for Extreme Classification with Application to Dynamic Search Advertising

The objective in extreme multi-label learning is to build classifiers that can annotate a data point with the subset of relevant labels from an extremely large label set. Extreme classification has, thus far, only been studied in the context of predicting labels for novel test points. This paper formulates the extreme classification problem when predictions need to be made on training points with partially revealed labels. This allows the reformulation of warm-start tagging, ranking and recommendation problems as extreme multi-label learning with each item to be ranked/recommended being mapped onto a separate label. The SwiftXML algorithm is developed to tackle such warm-start applications by leveraging label features. SwiftXML improves upon the state-of-the-art tree based extreme classifiers by partitioning tree nodes using two hyperplanes learnt jointly in the label and data point feature spaces. Optimization is carried out via an alternating minimization algorithm allowing SwiftXML to efficiently scale to large problems. Experiments on multiple benchmark tasks, including tagging on Wikipedia and item-to-item recommendation on Amazon, reveal that SwiftXML's predictions can be up to 14% more accurate as compared to leading extreme classifiers. SwiftXML also demonstrates the benefits of reformulating warm-start recommendation problems as extreme multi-label learning tasks by scaling beyond classical recommender systems and achieving prediction accuracy gains of up to 37%. Furthermore, in a live deployment for sponsored search on Bing, it was observed that SwiftXML could show significantly better ads as compared to a large ensemble of state-of-the-art techniques currently in production and could increase the relative click-through-rate by 10% while simultaneously reducing the bounce rate by 30%.

/pdf/extreme-multi-label-learning-with-label-features-for-warm-2309hj53nw.pdf

Yashoteja Prabhu

Papers

FastXML: a fast, accurate and stable tree-classifier for extreme multi-label learning

Extreme Multi-label Loss Functions for Recommendation, Tagging, Ranking & Other Missing Label Applications

Multi-label learning with millions of labels: recommending advertiser bid phrases for web pages

Parabel: Partitioned Label Trees for Extreme Classification with Application to Dynamic Search Advertising

Extreme Multi-label Learning with Label Features for Warm-start Tagging, Ranking & Recommendation