Technical and fundamental analysis are traditional tools used to analyze individual stocks; however, the finance literature has shown that the price movement of each individual stock correlates heavily with other stocks, especially those within the same sector. In this paper we propose a general-purpose market representation that incorporates fundamental and technical indicators and relationships between individual stocks. We treat the daily stock market as a ‘market image’ where rows (grouped by market sector) represent individual stocks and columns represent indicators. We apply a convolutional neural network over this market image to build market features in a hierarchical way. We use a recurrent neural network, with an attention mechanism over the market feature maps, to model temporal dynamics in the market. We show that our proposed model outperforms strong baselines in both short-term and long-term stock return prediction tasks. We also show another use for our market image: to construct concise and dense market embeddings suitable for downstream prediction tasks.

This year's conference broke a number of traditions set in America, HCOMP 2018 returned to Europe, where the very first HCOMP workshop had taken place in 2009. Besmira Nushi, Ece Kamar, and Eric interdisciplinary communities, we fostered new connections Horvitz were also singled out with an honorable among collective intelligence, crowdsourcing, mention for their paper "Towards Accountable AI: and human computation scholars and practitioners, Hybrid Human-Machine Analyses for Characterizing across diverse fields including humancomputer System Failure." Finally, Vikram Mohanty, David interaction (HCI), artificial intelligence, Thames, and Kurt Luther's presentation, "Are 1,000 economics, business, and design. Features Worth A Picture? Combining Crowdsourcing HCOMP was started by researchers from diverse and Face Recognition to Identify Civil War Soldiers," fields who wanted a high-quality scholarly venue for was given the Best Poster / Demo Presentation the review and presentation of the highest quality award. For this, we invited previous AAAI HCOMP conferences (and four submissions to a Works-in-Progress (WIP) and HCOMP workshops before that) to promote the most Demonstrations track, co-organized by Alessandro rigorous and exciting scholarship in this fast-emerging, Bozzon (Delft University of Technology) and Matteo multidisciplinary area.

Our work uses the notion of theory of mind to enable an interactive agent to keep track of the state of knowledge, goals and intentions of the human user, and to engage in and initiate sophisticated interactive behaviors using decision-theoretic paradigm of maximizing expected utility. Currently, systems like Google Now and Siri mostly react to user’s requests and commands using hand-crafted responses, but they cannot initiate intelligent communication and plan for longer term interactions. The reason is that they lack a clearly defined general objective of the interaction. Our main premise is that communication and interaction are types of action, so planning for communicative and interactive actions should be based on a unified framework of decisiontheoretic planning. To facilitate this, the system’s state of knowledge (a mental model) about the world has to include probabilistic representation of what is known, what is uncertain, and how things change as different events transpire. Further, the state of user’s knowledge and intentions (the theory of the user’s mind) needs to include precise specification of what the system knows, and how uncertain it is, about the user’s mental model, and about her desires and intentions. The theories of mind may be further nested to form interactive beliefs. Finally, decision-theoretic planning proposes that desirability of possible sequences of interactive and communicative actions be assessed as expected utilities of alternative plans.We describe our preliminary implementation using the Open CYC system, called MARTHA, and illustrate it in action using two simple interactive scenarios.

Deep neural networks have proven remarkably effective at solving many classification problems, but have been criticized recently for two major weaknesses: the reasons behind their predictions are uninterpretable, and the predictions themselves can often be fooled by small adversarial perturbations. These problems pose major obstacles for the adoption of neural networks in domains that require security or transparency. In this work, we evaluate the effectiveness of defenses that differentiably penalize the degree to which small changes in inputs can alter model predictions. Across multiple attacks, architectures, defenses, and datasets, we find that neural networks trained with this input gradient regularization exhibit robustness to transferred adversarial examples generated to fool all of the other models. We also find that adversarial examples generated to fool gradient-regularized models fool all other models equally well, and actually lead to more "legitimate," interpretable misclassifications as rated by people (which we confirm in a human subject experiment). Finally, we demonstrate that regularizing input gradients makes them more naturally interpretable as rationales for model predictions. We conclude by discussing this relationship between interpretability and robustness in deep neural networks.

The randomized-feature approach has been successfully employed in large-scale kernel approximation and supervised learning. The distribution from which the random features are drawn impacts the number of features required to efficiently perform a learning task. Recently, it has been shown that employing data-dependent randomization improves the performance in terms of the required number of random features. In this paper, we are concerned with the randomized-feature approach in supervised learning for good generalizability. We propose the Energy-based Exploration of Random Features (EERF) algorithm based on a data-dependent score function that explores the set of possible features and exploits the promising regions. We prove that the proposed score function with high probability recovers the spectrum of the best fit within the model class. Our empirical results on several benchmark datasets further verify that our method requires smaller number of random features to achieve a certain generalization error compared to the state-of-the-art while introducing negligible pre-processing overhead. EERF can be implemented in a few lines of code and requires no additional tuning parameters.

We study the problem of incentivizing high quality contributions in user generated content platforms, in which users arrive sequentially with unknown quality. We are interested in designing a content displaying strategy which decides which content should be chosen to show to users, with the goal of maximizing user experience (i.e., the likelihood of users liking the content).This goal naturally leads to a joint problem of incentivizing high quality contributions and learning the unknown content quality. To address the incentive issue, we consider a model in which users are strategic in deciding whether to contribute and are motivated by exposure, i.e., they aim to maximize the number of times their contributions are viewed. For the learning perspective, we model the content quality as the probability of obtaining positive feedback (e.g., like or upvote) from a random user. Naturally, the platform needs to resolve the classical trade-off between exploration (collecting feedback for all content) and exploitation (displaying the best content). We formulate this problem as a multi-arm bandit problem, where the number of arms (i.e., contributions) is increasing over time and depends on the strategic choices of arriving users. We first show that applying standard bandit algorithms incentivizes a flood of low cost contributions, which in turn leads to linear regret. We then propose Rand_UCB which adds an additional layer of randomization on top of the UCB algorithm to address the issue of flooding contributions. We show that Rand_UCB helps eliminate the incentives for low quality contributions, provides incentives for high quality contributions (due to bounded number of explorations for the low quality ones), and achieves sub-linear regrets with respect to displaying the current best arms.

The lack of interpretability remains a key barrier to the adoption of deep models in many applications. In this work, we explicitly regularize deep models so human users might step through the process behind their predictions in little time. Specifically, we train deep time-series models so their class-probability predictions have high accuracy while being closely modeled by decision trees with few nodes. Using intuitive toy examples as well as medical tasks for treating sepsis and HIV, we demonstrate that this new tree regularization yields models that are easier for humans to simulate than simpler L1 or L2 penalties without sacrificing predictive power.

We study a problem of optimal information gathering from multiple data providers that need to be incentivized to provide accurate information. This problem arises in many real world applications that rely on crowdsourced data sets, but where the process of obtaining data is costly. A notable example of such a scenario is crowd sensing. To this end, we formulate the problem of optimal information gathering as maximization of a submodular function under a budget constraint, where the budget represents the total expected payment to data providers. Contrary to the existing approaches, we base our payments on incentives for accuracy and truthfulness, in particular, peer prediction methods that score each of the selected data providers against its best peer, while ensuring that the minimum expected payment is above a given threshold. We first show that the problem at hand is hard to approximate within a constant factor that is not dependent on the properties of the payment function. However, for given topological and analytical properties of the instance, we construct two greedy algorithms, respectively called PPCGreedy and PPCGreedyIter, and establish theoretical bounds on their performance w.r.t. the optimal solution. Finally, we evaluate our methods using a realistic crowd sensing testbed.

We study minimal single-task peer prediction mechanisms that have limited knowledge about agents' beliefs. Without knowing what agents' beliefs are or eliciting additional information, it is not possible to design a truthful mechanism in a Bayesian-Nash sense. We go beyond truthfulness and explore equilibrium strategy profiles that are only partially truthful. Using the results from the multi-armed bandit literature, we give a characterization of how inefficient these equilibria are comparing to truthful reporting. We measure the inefficiency of such strategies by counting the number of dishonest reports that any minimal knowledge-bounded mechanism must have. We show that the order of this number is θ(log n), where n is the number of agents, and we provide a peer prediction mechanism that achieves this bound in expectation.

Reports on the 2017 AAAI Spring Symposium Series