Applying Deep Learning to Airbnb Search

Haldar, Malay; Abdool, Mustafa; Ramanathan, P.; Xu, Tao; Yang, Shulin; Duan, Huizhong; Zhang, Qing; Barrow-Williams, Nick; Turnbull, Bradley; Collins, Brendan M.; Legrand, Thomas

doi:10.1145/3292500.3330658

Cited by 74 publications

(51 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For some models such as the autoencoder, we found that log based transformations are helpful in improving performance and speeding up training. This type of behavior for neural network models has been reported previously [17]. For ease in explanation, we denote the set A to be the indices of the six features that are the most important features in our pricing algorithm, e.g., price, cost, average historical price.…”

Section: Featuresmentioning

confidence: 81%

Anomaly Detection for an E-commerce Pricing System

Ramakrishnan

Shaabani

2019

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

Online retailers execute a very large number of price updates when compared to brick-and-mortar stores. Even a few mis-priced items can have a significant business impact and result in a loss of customer trust. Early detection of anomalies in an automated real-time fashion is an important part of such a pricing system. In this paper, we describe unsupervised and supervised anomaly detection approaches we developed and deployed for a large-scale online pricing system at Walmart. Our system detects anomalies both in batch and real-time streaming settings, and the items flagged are reviewed and actioned based on priority and business impact. We found that having the right architecture design was critical to facilitate model performance at scale, and business impact and speed were important factors influencing model selection, parameter choice, and prioritization in a production environment for a large-scale system. We conducted analyses on the performance of various approaches on a test set using real-world retail data and fully deployed our approach into production. We found that our approach was able to detect the most important anomalies with high precision.

show abstract

Section: Featuresmentioning

confidence: 81%

Anomaly Detection for an E-commerce Pricing System

Ramakrishnan

Shaabani

2019

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

show abstract

“…However, the Mart-based models consume more memory. Nevertheless, uRank, uBoost, and urBoost models would have an advantage over Mart-based models by seamlessly linking to diverse deep learning embedding techniques when plain text features are present [19,29]. An interesting future direction would be leveraging the strengths of neural networks and trees.…”

Section: Resultsmentioning

confidence: 99%

Listwise Learning to Rank by Exploring Unique Ratings

Zhu

Klabjan

2020

Proceedings of the 13th International Conference on Web Search and Data Mining

View full text Add to dashboard Cite

In this paper, we propose new listwise learning-to-rank models that mitigate the shortcomings of existing ones. Existing listwise learning-to-rank models are generally derived from the classical Plackett-Luce model, which has three major limitations. (1) Its permutation probabilities overlook ties, i.e., a situation when more than one document has the same rating with respect to a query. This can lead to imprecise permutation probabilities and inefficient training because of selecting documents one by one. (2) It does not favor documents having high relevance.(3) It has a loose assumption that sampling documents at different steps is independent. To overcome the first two limitations, we model ranking as selecting documents from a candidate set based on unique rating levels in decreasing order. The number of steps in training is determined by the number of unique rating levels. More specifically, in each step, we apply multiple multi-class classification tasks to a document candidate set and choose all documents that have the highest rating from the document set. This is in contrast to taking one document step by step in the classical Plackett-Luce model. Afterward, we remove all of the selected documents from the document set and repeat until the remaining documents all have the lowest rating. We propose a new loss function and associated four models for the entire sequence of weighted classification tasks by assigning high weights to the selected documents with high ratings for optimizing Normalized Discounted Cumulative Gain (NDCG). To overcome the final limitation, we further propose a novel and efficient way of refining prediction scores by combining an adapted Vanilla Recurrent Neural Network (RNN) model with pooling given selected documents at previous steps. We encode all of the documents already selected by an RNN model. In a single step, we rank all of the documents with the same ratings using the last cell of the RNN multiple times. We have implemented our models using three settings: neural networks, neural networks with gradient boosting, and regression trees with gradient boosting. We have conducted experiments on four public datasets. The experiments demonstrate that the models notably outperform state-of-the-art learning-to-rank models.

show abstract

“…Organizational resources are becoming increasingly common across organizations as teams rely more heavily on ML [29,30,75]. We demonstrate that leveraging these auxiliary resources provides opportunity to train better models for related tasks of new modalities.…”

Section: Feature Generationmentioning

confidence: 99%

Leveraging organizational resources to adapt models to new data modalities

et al. 2020

View full text Add to dashboard Cite

As applications in large organizations evolve, the machine learning (ML) models that power them must adapt the same predictive tasks to newly arising data modalities (e.g., a new video content launch in a social media application requires existing text or image models to extend to video). To solve this problem, organizations typically create ML pipelines from scratch. However, this fails to utilize the domain expertise and data they have cultivated from developing tasks for existing modalities. We demonstrate how organizational resources, in the form of aggregate statistics, knowledge bases, and existing services that operate over related tasks, enable teams to construct a common feature space that connects new and existing data modalities. This allows teams to apply methods for data curation (e.g., weak supervision and label propagation) and model training (e.g., forms of multimodal learning) across these different data modalities. We study how this use of organizational resources composes at production scale in over 5 classification tasks at Google, and demonstrate how it reduces the time needed to develop models for new modalities from months to weeks or days.

show abstract

Applying Deep Learning to Airbnb Search

Cited by 74 publications

References 23 publications

Anomaly Detection for an E-commerce Pricing System

Anomaly Detection for an E-commerce Pricing System

Listwise Learning to Rank by Exploring Unique Ratings

Leveraging organizational resources to adapt models to new data modalities

Contact Info

Product

Resources

About