Learning mutually-grounded vision-language knowledge is a foundational task for cognitive systems and human-level artificial intelligence. Most of knowledge-learning techniques are focused on single modal representations in a static environment with a fixed set of data. Here, we explore an ecologically more-plausible setting by using a stream of cartoon videos to build vision-language concept hierarchies continuously. This approach is motivated by the literature on cognitive development in early childhood. We present the model of deep concept hierarchy (DCH) that enables the progressive abstraction of concept knowledge in multiple levels. We develop a stochastic method for graph construction, i.e. a graph Monte Carlo algorithm, to search efficiently the huge compositional space of the vision-language concepts. The concept hierarchies are built incrementally and can handle concept drift, allowing for being deployed in lifelong learning environments. Using a series of approximately 200 episodes of educational cartoon videos we demonstrate the emergence and evolution of the concept hierarchies as the video stories unfold. We also present the application of the deep concept hierarchies for context-dependent translation between vision and language, i.e. the transcription of a visual scene into text and the generation of visual imagery from text.

Error backpropagation is an extremely effective algorithm for assigning credit in artificial neural networks. However, weight updates under Backprop depend on lengthy recursive computations and require separate output and error messages — features not shared by biological neurons, that are perhaps unnecessary. In this paper, we revisit Backprop and the credit assignment problem. We first decompose Backprop into a collection of interacting learning algorithms; provide regret bounds on the performance of these sub-algorithms; and factorize Backprop's error signals. Using these results, we derive a new credit assignment algorithm for nonparametric regression, Kickback, that is significantly simpler than Backprop. Finally, we provide a sufficient condition for Kickback to follow error gradients, and show that Kickback matches Backprop's performance on real-world regression benchmarks.

With the enormous scale of massive open online courses (MOOCs), peer grading is vital for addressing the assessment challenge for open-ended assignments or exams while at the same time providing students with an effective learning experience through involvement in the grading process. Most existing MOOC platforms use simple schemes for aggregating peer grades, e.g., taking the median or mean. To enhance these schemes, some recent research attempts have developed machine learning methods under either the cardinal setting (for absolute judgment) or the ordinal setting (for relative judgment). In this paper, we seek to study both cardinal and ordinal aspects of peer grading within a common framework. First, we propose novel extensions to some existing probabilistic graphical models for cardi- nal peer grading. Not only do these extensions give su- perior performance in cardinal evaluation, but they also outperform conventional ordinal models in ordinal eval- uation. Next, we combine cardinal and ordinal models by augmenting ordinal models with cardinal predictions as prior. Such combination can achieve further performance boosts in both cardinal and ordinal evaluations, suggesting a new research direction to pursue for peer grading on MOOCs. Extensive experiments have been conducted using real peer grading data from a course called “Science, Technology, and Society in China I” offered by HKUST on the Coursera platform.

Typically a user prefers an item (e.g., a movie) because she likes certain features of the item (e.g., director, genre, producer). This observation motivates us to consider a feature-centric recommendation approach to item recommendation: instead of directly predicting the rating on items, we predict the rating on the features of items, and use such ratings to derive the rating on an item. This approach offers several advantages over the traditional item-centric approach: it incorporates more information about why a user chooses an item, it generalizes better due to the denser feature rating data, it explains the prediction of item ratings through the predicted feature ratings. Another contribution is turning a principled item-centric solution into a feature-centric solution, instead of inventing a new algorithm that is feature-centric. This approach maximally leverages previous research. We demonstrate this approach by turning the traditional item-centric latent factor model into a feature-centric solution and demonstrate its superiority over item-centric approaches.

With the rapid development of signed social networks in which therelationships between two nodes can be either positive (indicatingrelations such as like) or negative (indicating relations such asdislike), producing a personalized ranking list with positive linkson the top and negative links at the bottom is becoming anincreasingly important task. To accomplish it, we propose ageneralized AUC (GAUC) to quantify the ranking performance ofpotential links (including positive, negative, and unknown statuslinks) in partially observed signed social networks. In addition, wedevelop a novel link recommendation algorithm by directly optimizingthe GAUC loss. We conduct experimental studies based upon Wikipedia,MovieLens, and Slashdot; our results demonstrate the effectivenessand the efficiency of the proposed approach.

Recent work has demonstrated the value of social media monitoring for health surveillance (e.g., tracking influenza or depression rates). It is an open question whether such data can be used to make causal inferences (e.g., determining which activities lead to increased depression rates). Even in traditional, restricted domains, estimating causal effects from observational data is highly susceptible to confounding bias. In this work, we estimate the effect of exercise on mental health from Twitter, relying on statistical matching methods to reduce confounding bias. We train a text classifier to estimate the volume of a user's tweets expressing anxiety, depression, or anger, then compare two groups: those who exercise regularly (identified by their use of physical activity trackers like Nike+), and a matched control group. We find that those who exercise regularly have significantly fewer tweets expressing depression or anxiety; there is no significant difference in rates of tweets expressing anger. We additionally perform a sensitivity analysis to investigate how the many experimental design choices in such a study impact the final conclusions, including the quality of the classifier and the construction of the control group.

Understanding the demographics of users of online social networks has important applications for health, marketing, and public messaging. In this paper, we predict the demographics of Twitter users based on whom they follow. Whereas most prior approaches rely on a supervised learning approach, in which individual users are labeled with demographics, we instead create a distantly labeled dataset by collecting audience measurement data for 1,500 websites (e.g., 50% of visitors to gizmodo.com are estimated to have a bachelor's degree). We then fit a regression model to predict these demographics using information about the followers of each website on Twitter. The resulting average held-out correlation is .77 across six different variables (gender, age, ethnicity, education, income, and child status). We additionally validate the model on a smaller set of Twitter users labeled individually for ethnicity and gender, finding performance that is surprisingly competitive with a fully supervised approach.

In this paper, we study protocols that allow to discern conscious and unconscious decisions of human beings; i.e., protocols that measure awareness. Consciousness is a central research theme in Neuroscience and AI, which remains, to date, an obscure phenomenon of human brains. Our starting point is a recent experiment, called Post Decision Wagering (PDW) (Persaud, McLeod, and Cowey 2007), that attempts to align experimenters' and subjects' objectives by leveraging financial incentives. We note a similarity with mechanism design, a research area which aims at the design of protocols that reconcile often divergent objectives through incentive-compatibility. We look at the issue of measuring awareness from this perspective. We abstract the setting underlying the PDW experiment and identify three factors that could make it ineffective: rationality, risk attitude and bias of subjects. Using mechanism design tools, we study the barrier between possibility and impossibility of incentive compatibility with respect to the aforementioned characteristics of subjects. We complete this study by showing how to use our mechanisms to potentially get a better understanding of consciousness.

Online advertising is an important and huge industry. Having knowledge of the website attributes can contribute greatly to business strategies for ad-targeting, content display, inventory purchase or revenue prediction. In this paper, we introduce a stochastic blockmodeling for the website relations induced by the event of online user visitation. We propose two clustering algorithms to discover the intrinsic structures of websites, and compare the performance with a goodness-of-fit method and a deterministic graph partitioning method. We demonstrate the effectiveness of our algorithms on both simulation and AOL website dataset.

Reinforcement learning for robot control tasks in continuous environments is a challenging problem due to the dimensionality of the state and action spaces, time and resource costs for learning with a real robot as well as constraints imposed for its safe operation. In this paper we propose a model-based reinforcement learning approach for continuous environments with constraints. The approach combines model-based reinforcement learning with recent advances in approximate optimal control. This results in a bounded-rationality agent that makes decisions in real-time by efficiently solving a sequence of constrained optimization problems on learned sparse Gaussian process models. Such a combination has several advantages. No high-dimensional policy needs to be computed or stored while the learning problem often reduces to a set of lower-dimensional models of the dynamics. In addition, hard constraints can easily be included and objectives can also be changed in real-time to allow for multiple or dynamic tasks. The efficacy of the approach is demonstrated on both an extended cart pole domain and a challenging quadcopter navigation task using real data.