Incorporating Behavioral Constraints in Online AI Systems

AI systems that learn through reward feedback about the actions they take are increasingly deployed in domains that have significant impact on our daily life. However, in many cases the online rewards should not be the only guiding criteria, as there are additional constraints and/or priorities imposed by regulations, values, preferences, or ethical principles. We detail a novel online agent that learns a set of behavioral constraints by observation and uses these learned constraints as a guide when making decisions in an online setting while still being reactive to reward feedback. To define this agent, we propose to adopt a novel extension to the classical contextual multi-armed bandit setting and we provide a new algorithm called Behavior Constrained Thompson Sampling (BCTS) that allows for online learning while obeying exogenous constraints. Our agent learns a constrained policy that implements the observed behavioral constraints demonstrated by a teacher agent, and then uses this constrained policy to guide the reward-based online exploration and exploitation. We characterize the upper bound on the expected regret of the contextual bandit algorithm that underlies our agent and provide a case study with real world data in two application domains. Our experiments show that the designed agent is able to act within the set of behavior constraints without significantly degrading its overall reward performance.

Adaptive Representation Selection in Contextual Bandit

We consider an extension of the contextual bandit setting, motivated by several practical applications, where an unlabeled history of contexts can become available for pre-training before the online decision-making begins. We propose an approach for improving the performance of contextual bandit in such setting, via adaptive, dynamic representation learning, which combines offline pre-training on unlabeled history of contexts with online selection and modification of embedding functions. Our experiments on a variety of datasets and in different nonstationary environments demonstrate clear advantages of our approach over the standard contextual bandit.

Scalable Recollections for Continual Lifelong Learning

Given the recent success of Deep Learning applied to a variety of single tasks, it is natural to consider more human-realistic settings. Perhaps the most difficult of these settings is that of continual lifelong learning, where the model must learn online over a continuous stream of non-stationary data. A continual lifelong learning system must have three primary capabilities to succeed: it must learn and adapt over time, it must not forget what it has learned, and it must be efficient in both training time and memory. Recent techniques have focused their efforts largely on the first two capabilities while the third capability remains largely unexplored. In this paper, we consider the problem of efficient and effective storage of experiences over very large time-frames. In particular we consider the case where typical experiences are n bits and memories are limited to k bits for k << n. We present a novel scalable architecture and training algorithm in this challenging domain and provide an extensive evaluation of its performance. Our results show that we can achieve considerable gains on top of state-of-the-art methods such as GEM.

Context Attentive Bandits: Contextual Bandit with Restricted Context

We consider a novel formulation of the multi-armed bandit model, which we call the contextual bandit with restricted context, where only a limited number of features can be accessed by the learner at every iteration. This novel formulation is motivated by different online problems arising in clinical trials, recommender systems and attention modeling. Herein, we adapt the standard multi-armed bandit algorithm known as Thompson Sampling to take advantage of our restricted context setting, and propose two novel algorithms, called the Thompson Sampling with Restricted Context(TSRC) and the Windows Thompson Sampling with Restricted Context(WTSRC), for handling stationary and nonstationary environments, respectively. Our empirical results demonstrate advantages of the proposed approaches on several real-life datasets

Bandit Models of Human Behavior: Reward Processing in Mental Disorders

Drawing an inspiration from behavioral studies of human decision making, we propose here a general parametric framework for multi-armed bandit problem, which extends the standard Thompson Sampling approach to incorporate reward processing biases associated with several neurological and psychiatric conditions, including Parkinson's and Alzheimer's diseases, attention-deficit/hyperactivity disorder (ADHD), addiction, and chronic pain. We demonstrate empirically that the proposed parametric approach can often outperform the baseline Thompson Sampling on a variety of datasets. Moreover, from the behavioral modeling perspective, our parametric framework can be viewed as a first step towards a unifying computational model capturing reward processing abnormalities across multiple mental conditions.

Multi-armed Bandit Problem with Known Trend

We consider a variant of the multi-armed bandit model, which we call multi-armed bandit problem with known trend, where the gambler knows the shape of the reward function of each arm but not its distribution. This new problem is motivated by different online problems like active learning, music and interface recommendation applications, where when an arm is sampled by the model the received reward change according to a known trend. By adapting the standard multi-armed bandit algorithm UCB1 to take advantage of this setting, we propose the new algorithm named A-UCB that assumes a stochastic model. We provide upper bounds of the regret which compare favourably with the ones of UCB1. We also confirm that experimentally with different simulations

Freshness-Aware Thompson Sampling

To follow the dynamicity of the user's content, researchers have recently started to model interactions between users and the Context-Aware Recommender Systems (CARS) as a bandit problem where the system needs to deal with exploration and exploitation dilemma. In this sense, we propose to study the freshness of the user's content in CARS through the bandit problem. We introduce in this paper an algorithm named Freshness-Aware Thompson Sampling (FA-TS) that manages the recommendation of fresh document according to the user's risk of the situation. The intensive evaluation and the detailed analysis of the experimental results reveals several important discoveries in the exploration/exploitation (exr/exp) behaviour.

A Neural Networks Committee for the Contextual Bandit Problem

This paper presents a new contextual bandit algorithm, NeuralBandit, which does not need hypothesis on stationarity of contexts and rewards. Several neural networks are trained to modelize the value of rewards knowing the context. Two variants, based on multi-experts approach, are proposed to choose online the parameters of multi-layer perceptrons. The proposed algorithms are successfully tested on a large dataset with and without stationarity of rewards.

Exponentiated Gradient Exploration for Active Learning

Active learning strategies respond to the costly labelling task in a supervised classification by selecting the most useful unlabelled examples in training a predictive model. Many conventional active learning algorithms focus on refining the decision boundary, rather than exploring new regions that can be more informative. In this setting, we propose a sequential algorithm named EG-Active that can improve any Active learning algorithm by an optimal random exploration. Experimental results show a statistically significant and appreciable improvement in the performance of our new approach over the existing active feedback methods.

R-UCB: a Contextual Bandit Algorithm for Risk-Aware Recommender Systems

Mobile Context-Aware Recommender Systems can be naturally modelled as an exploration/exploitation trade-off (exr/exp) problem, where the system has to choose between maximizing its expected rewards dealing with its current knowledge (exploitation) and learning more about the unknown user's preferences to improve its knowledge (exploration). This problem has been addressed by the reinforcement learning community but they do not consider the risk level of the current user's situation, where it may be dangerous to recommend items the user may not desire in her current situation if the risk level is high. We introduce in this paper an algorithm named R-UCB that considers the risk level of the user's situation to adaptively balance between exr and exp. The detailed analysis of the experimental results reveals several important discoveries in the exr/exp behaviour.