This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn useful representations, we systematically study the major components of our framework. We show that (1) composition of data augmentations plays a critical role in defining effective predictive tasks, (2) introducing a learnable nonlinear transformation between the representation and the contrastive loss substantially improves the quality of the learned representations, and (3) contrastive learning benefits from larger batch sizes and more training steps compared to supervised learning. By combining these findings, we are able to considerably outperform previous methods for self-supervised and semi-supervised learning on ImageNet. A linear classifier trained on self-supervised representations learned by SimCLR achieves 76.5% top-1 accuracy, which is a 7% relative improvement over previous state-of-the-art, matching the performance of a supervised ResNet-50. When fine-tuned on only 1% of the labels, we achieve 85.8% top-5 accuracy, outperforming AlexNet with 100X fewer labels.

A Simple Framework for Contrastive Learning of Visual Representations

Even though human movement and mobility patterns have a high degree of freedom and variation, they also exhibit structural patterns due to geographic and social constraints. Using cell phone location data, as well as data from two online location-based social networks, we aim to understand what basic laws govern human motion and dynamics. We find that humans experience a combination of periodic movement that is geographically limited and seemingly random jumps correlated with their social networks. Short-ranged travel is periodic both spatially and temporally and not effected by the social network structure, while long-distance travel is more influenced by social network ties. We show that social relationships can explain about 10% to 30% of all human movement, while periodic behavior explains 50% to 70%. Based on our findings, we develop a model of human mobility that combines periodic short range movements with travel due to the social network structure. We show that our model reliably predicts the locations and dynamics of future human movement and gives an order of magnitude better performance than present models of human mobility.

/pdf/friendship-and-mobility-user-movement-in-location-based-2u0rrxuc9l.pdf

Friendship and mobility: user movement in location-based social networks

YouTube represents one of the largest scale and most sophisticated industrial recommendation systems in existence. In this paper, we describe the system at a high level and focus on the dramatic performance improvements brought by deep learning. The paper is split according to the classic two-stage information retrieval dichotomy: first, we detail a deep candidate generation model and then describe a separate deep ranking model. We also provide practical lessons and insights derived from designing, iterating and maintaining a massive recommendation system with enormous user-facing impact.

https://dl.acm.org/doi/pdf/10.1145/2959100.2959190

Deep Neural Networks for YouTube Recommendations

Building a successful recommender system depends on understanding both the dimensions of people's preferences as well as their dynamics. In certain domains, such as fashion, modeling such preferences can be incredibly difficult, due to the need to simultaneously model the visual appearance of products as well as their evolution over time. The subtle semantics and non-linear dynamics of fashion evolution raise unique challenges especially considering the sparsity and large scale of the underlying datasets. In this paper we build novel models for the One-Class Collaborative Filtering setting, where our goal is to estimate users' fashion-aware personalized ranking functions based on their past feedback. To uncover the complex and evolving visual factors that people consider when evaluating products, our method combines high-level visual features extracted from a deep convolutional neural network, users' past feedback, as well as evolving trends within the community. Experimentally we evaluate our method on two large real-world datasets from Amazon.com, where we show it to outperform state-of-the-art personalized ranking measures, and also use it to visualize the high-level fashion trends across the 11-year span of our dataset.

Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering

Twitter as a new form of social media can potentially contain much useful information, but content analysis on Twitter has not been well studied. In particular, it is not clear whether as an information source Twitter can be simply regarded as a faster news feed that covers mostly the same information as traditional news media. In This paper we empirically compare the content of Twitter with a traditional news medium, New York Times, using unsupervised topic modeling. We use a Twitter-LDA model to discover topics from a representative sample of the entire Twitter. We then use text mining techniques to compare these Twitter topics with topics from New York Times, taking into consideration topic categories and types. We also study the relation between the proportions of opinionated tweets and retweets and topic categories and types. Our comparisons show interesting and useful findings for downstream IR or DM applications.

/pdf/comparing-twitter-and-traditional-media-using-topic-models-40ux4xfa4w.pdf

Comparing twitter and traditional media using topic models

Social networks such as Facebook, LinkedIn, and Twitter have been a crucial source of information for a wide spectrum of users. In Twitter, popular information that is deemed important by the community propagates through the network. Studying the characteristics of content in the messages becomes important for a number of tasks, such as breaking news detection, personalized message recommendation, friends recommendation, sentiment analysis and others. While many researchers wish to use standard text mining tools to understand messages on Twitter, the restricted length of those messages prevents them from being employed to their full potential.We address the problem of using standard topic models in micro-blogging environments by studying how the models can be trained on the dataset. We propose several schemes to train a standard topic model and compare their quality and effectiveness through a set of carefully designed experiments from both qualitative and quantitative perspectives. We show that by training a topic model on aggregated messages we can obtain a higher quality of learned model which results in significantly better performance in two real-world classification problems. We also discuss how the state-of-the-art Author-Topic model fails to model hierarchical relationships between entities in Social Media.

/pdf/empirical-study-of-topic-modeling-in-twitter-12cqhvfswo.pdf

Empirical study of topic modeling in Twitter

Social network services have become a viable source of information for users. In Twitter, information deemed important by the community propagates through retweets. Studying the characteristics of such popular messages is important for a number of tasks, such as breaking news detection, personalized message recommendation, viral marketing and others. This paper investigates the problem of predicting the popularity of messages as measured by the number of future retweets and sheds some light on what kinds of factors influence information propagation in Twitter. We formulate the task into a classification problem and study two of its variants by investigating a wide spectrum of features based on the content of the messages, temporal information, metadata of messages and users, as well as structural properties of the users' social graph on a large scale dataset. We show that our method can successfully predict messages which will attract thousands of retweets with good performance.

/pdf/predicting-popular-messages-in-twitter-4eax7xnf3u.pdf

Predicting popular messages in Twitter

Micro-blogging services have become indispensable communication tools for online users for disseminating breaking news, eyewitness accounts, individual expression, and protest groups. Recently, Twitter, along with other online social networking services such as Foursquare, Gowalla, Facebook and Yelp, have started supporting location services in their messages, either explicitly, by letting users choose their places, or implicitly, by enabling geo-tagging, which is to associate messages with latitudes and longitudes. This functionality allows researchers to address an exciting set of questions: 1) How is information created and shared across geographical locations, 2) How do spatial and linguistic characteristics of people vary across regions, and 3) How to model human mobility. Although many attempts have been made for tackling these problems, previous methods are either complicated to be implemented or oversimplified that cannot yield reasonable performance. It is a challenge task to discover topics and identify users' interests from these geo-tagged messages due to the sheer amount of data and diversity of language variations used on these location sharing services. In this paper we focus on Twitter and present an algorithm by modeling diversity in tweets based on topical diversity, geographical diversity, and an interest distribution of the user. Furthermore, we take the Markovian nature of a user's location into account. Our model exploits sparse factorial coding of the attributes, thus allowing us to deal with a large and diverse set of covariates efficiently. Our approach is vital for applications such as user profiling, content recommendation and topic tracking. We show high accuracy in location estimation based on our model. Moreover, the algorithm identifies interesting topics based on location and language.

/pdf/discovering-geographical-topics-in-the-twitter-stream-3smovo4y25.pdf

Discovering geographical topics in the twitter stream

Web 2.0 has led to the development and evolution of web-based communities and applications. These communities provide places for information sharing and collaboration. They also open t he door for inappropriate online activities, such as harassment, i n which some users post messages in a virtual community that are intention- ally offensive to other members of the community. It is a new and challenging task to detect online harassment; currently fe w systems attempt to solve this problem. In this paper, we use a supervised learning approach for dete ct- ing harassment. Our technique employs content features, sentiment features, and contextual features of documents. The experi mental results described herein show that our method achieves significant improvements over several baselines, including Term Frequency- Inverse Document Frequency (TFIDF) approaches. Identifica tion of online harassment is feasible when TFIDF is supplemented with sentiment and contextual feature attributes.

/pdf/detection-of-harassment-on-web-2-0-38xklkierx.pdf

Detection of Harassment on Web 2.0

Many internet companies, such as Yahoo, Facebook, Google and Twitter, rely on content recommendation systems to deliver the most relevant content items to individual users through personalization. Delivering such personalized user experiences is believed to increase the long term engagement of users. While there has been a lot of progress in designing effective personalized recommender systems, by exploiting user interests and historical interaction data through implicit (item click) or explicit (item rating) feedback, directly optimizing for users' satisfaction with the system remains challenging. In this paper, we explore the idea of using item-level dwell time as a proxy to quantify how likely a content item is relevant to a particular user. We describe a novel method to compute accurate dwell time based on client-side and server-side logging and demonstrate how to normalize dwell time across different devices and contexts. In addition, we describe our experiments in incorporating dwell time into state-of-the-art learning to rank techniques and collaborative filtering models that obtain competitive performances in both offline and online settings.

/pdf/beyond-clicks-dwell-time-for-personalization-1bworbfyzx.pdf

Liangjie Hong

Papers

Empirical study of topic modeling in Twitter

Predicting popular messages in Twitter

Discovering geographical topics in the twitter stream

Detection of Harassment on Web 2.0

Beyond clicks: dwell time for personalization