In this work, we present a novel human-in-the-loop framework to help the agent understand the decision making process that involves choosing preferred options. We focus on qualitative preference models over alternatives from combinatorial domains. This framework is interactive: e.g., the agent provides her behavioral data to the framework, and the framework ex- plains the learned model to the agent. It is iterative: the framework collects feedback on the learned model from the agent and tries to improve it accordingly until the agent terminates the iteration. In order to communicate the learned preference model to the agent, we focus on visualizing some of the intuitive and explain- able graphic models, such as lexicographic preference trees and forests, and conditional preference networks. To this end, we discuss key aspects of our framework, and demonstrate our prototype ready for lexicographic preference models.

Classic k-Nearest Neighbor (kNN) algorithms approximate a regression or classification function at a query point based on the k-nearest training observations. In real-world datasets, however, the set of k neighbors is frequently not uniformly distributed around a given query point. This can result in a locally biased estimate and thus in degraded regression or classification results. This paper presents two new kNN algorithms that adjust the weight of the k-nearest neighbors to achieve a more balanced distribution. Experiments on real-world datasets and a range of synthetic training distributions and noise levels identify conditions under which the algorithms can improve accuracy with minimal increase in computation time.

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.

The representation of semantic meaning of sentences using neural network has recently gained popularity, due to the fact that there is no need to specifically extract lexical syntactic and semantic features. A major problem with this approach is that it requires large human annotated corpora. In order to reduce human annotation effort, in recent years, researchers made several attempts to find universal sentence representation methods, aiming to obtain general-purpose sentence embeddings that could be widely adopted to a wide range of NLP tasks without training directly from the specific datasets. InferSent, a supervised universal sentence representation model proposed by Facebook research, implements 8 popular neural network sentence encoding structures trained on natural language inference datasets, and apply to 12 different NLP tasks. However, the relation classification task was not one of these. In this paper, we re-train these 8 sentence encoding structures and use them as the starting points on relation classification task. Experiments using SemEval-2010 datasets show that our models could achieve comparable results to the state-of-the-art relation classification systems.

Learning from Observation (LfO) is highly useful for modeling behaviors through nonintrusive observation of some actor's performance. However, an actor's performance is often influenced by unobservable internal influences, such as emotions, agendas, and memory of past events. Therefore, new techniques are needed to infer the structure of these influences and their effect on an actor's decisions. In this paper, we propose a novel approach called Memory Composition Learning (MCL) for capturing one internal influence: memory of past events. We hypothesize that memory influences on a behavior can be modeled through parameterized memory features that can be learned from observation of traces of an actor's behavior; these memory features can then be presented as additional input to a performance modeling application. We demonstrate the efficacy of our approach in a simulated vacuum cleaner domain and show that hidden memory influences can be detected, modeled, and then used to improve machine learning performance.

A crucial point for ensemble learning systems is the capacity of making different errors on any given sample, which highlights the importance of diversity for ensemble-based decision systems. A usual way of increasing diversity is to combine traditional ensemble methods. Based on this context, we propose a novel combining-based algorithm of pool generation using a merging of bagging, random patches, and boosting techniques for ensemble regression problems. Numerical results indicate that, depending on both the dataset and the diversity measurement, our proposal generates a pool of regressors with more diversity when compared to single ensemble generator approaches.

During opinion formation, interacting agents can be assumed to be engaging in learning and decision-making processes to satisfy their individual goals. These goals are determined by the agents’ preferences – which are often unknown, complex and unpredictable. Most opinion formation frameworks however, assume static preferences and fail to model practical situations where human preferences change. We propose a new framework to simulate the process of opinion formation under uncertainty and dynamism. Agents who are unaware of their implicit con-textual preferences utilize inverse reinforcement learning to compute reward functions that determines their preferences. Reinforcement learning is subsequently used to optimize the agents’ behavior and satisfy their individual goals. The novelty of our approach lies in its ability to capture uncertainty and dynamism in the agent’s preferences, which are assumed to be unknown initially. This framework is compared to a baseline method based on reinforcement learning, and results show its ability to per-form better under dynamic scenarios.

Classification problem in authorship attribution consists of choosing the correct author of a document from an exhaustive list of candidates presented by the samples of their writing. A typical approach is to assign a vector representing measurements of a stylometric feature to each sample document and apply a supervised machine learning method to build a classifier. Different classifiers vary in the accuracy and attributions of the disputed documents. In our previous research, we have shown that a large number of classifiers can be combined into an effective jury via weighted voting. Such a jury is almost always more accurate than individual classifiers. In this paper, we investigate whether it is possible to improve a jury’s accuracy by eliminating some of its members. We test and compare two methods of reduction. Dynamic reduction selects a subset of original jury members by eliminating sycophants. Static reduction tests the behavior of preselected juries. Our testbed is a collection of 18th-century political writings, a fertile research ground rich with disputed works.

A denial-of-service (DoS) attack is a malicious act with the goal of interrupting the access to a computer network. The result of DoS attack can cause the computers on the network to squander their resources to serve illegitimate requests that result in a disruption of the network’s services to legitimate users. With a sophisticated DoS attack, it becomes difficult to distinguish malicious requests from legitimate requests. Since a network layer DoS attack can cause interruptions to a network while causing collateral damage, it is vital to understand the measures to mitigate against such attacks. Generally, approaches that implement distribution charts based on statistical analysis or honeypots have been applied to detect a DoS attack. However, this is usually too late, as the damage is already done. We hypothesize in this work that a graph-based approach can provide the capability to identify a DoS attack at its inception. A graph-based approach will also allow us to not only focus on anomalies within an entity (like a computer) but also allow us to analyze the anomalies that exist in an entity’s relationship with other entities, thus providing a rich source of contextual analysis. We demonstrate our proposed approach using a publicly-available dataset.

Conditional independence structures describe independencies of one set of variables from another set of variables conditioned upon a third set of variables. These structures are invaluable means for compact representations of knowledge because independencies can be exploited for useful factorizations. Conditional independence structures appear in different disguise in various areas of knowledge representation, be it the conditional independence of sets of random variables in probabilistic graphical models such as Bayesian networks or as conditional functions related to belief revision, or as in- dependencies induced by (embedded) multivalued dependencies in data bases. This paper investigates conditional independencies for Boolean functions using Fourier analysis. We define three notions of independence based on the notion of influence of a variable on a function and draw connections to multivalued dependencies.