This paper presents a novel framework for online incremental place recognition in an indoor environment. The framework addresses the scenario in which scene images are gradually obtained during long-term operation in the real-world indoor environment. Multiple users may interact with the classification system and confirm either current or past prediction results; the system then immediately updates itself to improve the classification system. This framework is based on the proposed \emph{n}-value self-organizing and incremental neural network (\emph{n}-SOINN), which has been derived by modifying the original SOINN to be appropriate for use in scene recognition. The evaluation was performed on the standard MIT 67-category indoor scene dataset and shows that the proposed framework achieves the same accuracy as that of the state-of-the-art offline method, while the computation time of the proposed framework is significantly faster and fully incremental update is allowed. Additionally, a small extra set of training samples is incrementally given to the system to simulate the incremental learning situation. The result shows that the proposed framework can leverage such additional samples and achieve the state-of-the-art result.

Given a classification task, what is the best way to teach the resulting boundary to a human? While machine learning techniques can provide excellent methods for finding the boundary, including the selection of examples in an online setting, they tell us little about how we would teach a human the same task. We propose to investigate the problem of example selection and presentation in the context of teaching humans, and explore a variety of mechanisms in the interests of finding what may work best. In particular, we begin with the baseline of random presentation and then examine combinations of several mechanisms: the indication of an example’s relative difficulty, the use of the shaping heuristic from the cognitive science literature (moving from easier examples to harder ones), and a novel kernel-based “coverage model” of the subject’s mastery of the task. From our experiments on 54 human subjects learning and performing a pair of synthetic classification tasks via our teaching system, we found that we can achieve the greatest gains with a combination of shaping and the coverage model.

We study hybrid online-batch matching problems, where agents arrive continuously, but are only matched in periodic rounds, when many of them can be considered simultaneously. Agents not getting matched in a given round remain in the market for the next round. This setting models several scenarios of interest, including many job markets as well as kidney exchange mechanisms. We consider the social utility of two commonly used mechanisms for such markets: one that aims for stability in each round (greedy), and one that attempts to maximize social utility in each round (max-weight). Surprisingly, we find that in the long term, the social utility of the greedy mechanism can be higher than that of the max-weight mechanism. We hypothesize that this is because the greedy mechanism behaves similarly to a soft threshold mechanism, where all connections below a certain threshold are rejected by the participants in favor of waiting until the next round. Motivated by this observation, we propose a method to approximately calculate the optimal threshold for an individual agent to use based on characteristics of the other agents participating, and demonstrate experimentally that social utility is high when all agents use this strategy. Thresholding can also be applied by the mechanism itself to improve social welfare; we demonstrate this with an example on graphs that model pairwise kidney exchange.

Adaptive strategies are popular in poker AI research. Three desirable properties of an adaptive poker bot are optimality, generality and speed of response to new opponents. These three properties though cannot be achieved simultaneously for most imperfect information games; some trade-off must be made between them. This general principle is connected to recent work on poker AI and particularly the total bankroll competitions at the Annual Computer Poker Competition (ACPC). This paper is meant to generate discussion between researchers about how to explain and quantify these trade-offs better and to possible future directions for research.

We describe a game environment we have developed that we call the Strategy Engine for Programming Intelligent Agents (SEPIA). SEPIA is based on real-time strategy games, but modified extensively to preferentially support the development of artificial agents rather than human play. Through flexible configuration options, SEPIA is designed to be pedagogically scalable: suitable for use at the undergraduate and graduate levels, and also as a research testbed. We also describe assignments and our experiences with this environment in undergraduate and graduate classes.

Manipulation, bribery, and control are well-studied ways of changing the outcome of an election. Many voting systems are, in the general case, computationally resistant to some of these manipulative actions. However when restricted to single-peaked electorates, these systems suddenly become easy to manipulate. Recently, Faliszewski, Hemaspaandra, and Hemaspaandra studied the complexity of dishonest behavior in nearly single-peaked electorates. These are electorates that are not single-peaked but close to it according to some distance measure. In this paper we introduce several new distance measures regarding single-peakedness. We prove that determining whether a given profile is nearly single-peaked is NP-complete in many cases. For one case we present a polynomial-time algorithm. Furthermore, we explore the relations between several notions of nearly single-peakedness.

Despite committees and elections being widespread in thereal-world, the design of agents for operating in humancomputer committees has received far less attention than thetheoretical analysis of voting strategies. We address this gapby providing an agent design that outperforms other voters ingroups comprising both people and computer agents. In oursetting participants vote by simultaneously submitting a ranking over a set of candidates and the election system uses a social welfare rule to select a ranking that minimizes disagreements with participants’ votes. We ran an extensive studyin which hundreds of people participated in repeated votingrounds with other people as well as computer agents that differed in how they employ strategic reasoning in their votingbehavior. Our results show that over time, people learn todeviate from truthful voting strategies, and use heuristics toguide their play, such as repeating their vote from the previous round. We show that a computer agent using a bestresponse voting strategy was able to outperform people in thegame. Our study has implication for agent designers, highlighting the types of strategies that enable agents to succeedin committees comprising both human and computer participants. This is the first work to study the role of computeragents in voting settings involving both human and agent participants.

In this paper we investigate the applications of Elo ratings (originally designed for 2-player chess) to a heterogeneous nonlinear multiagent system to determine an agent's overall impact on its team's performance. Measuring this impact has been attempted in many different ways, including reward shaping; the generation of heirarchies, holarchies, and teams; mechanism design; and the creation of subgoals. We show that in a multiagent system, an Elo rating will accurately reflect an agent's ability to contribute positively to a team's success with no need for any other feedback than a repeated binary win/loss signal. The Elo rating not only measures ``personal" success, but simultaneously success in assisting other agents to perform favorably.

The introduction of Electric Vehicles (EVs) in the existing Energy Grid raises many issues regarding Grid stability and charging behavior. Uncontrolled charging on the customer’s side may increase the already high peaks in the energy demand that lead to respective increase in the energy prices.We propose a novel smart charging algorithm that maximizes individual welfare and reduces the individual energy expenses. We use Reinforcement Learning trained on real world data to learn the individual household consumption behavior and propose a charging algorithm with respect to individual welfare maximization objective. Furthermore, we use statistical customer models to simulate the EV customer behavior. We show that the individual customers, represented by intelligent agents, using the proposed charging algorithm reduce their energy expenses. Additionally, we show that the average energy prices, on an aggregated level, are reduced as a result of smarter use of the energy available. Finally we prove that the presented algorithm achieves significant peak reduction and reshaping of the energy demand curve.

The PowerTAC competition focuses on trading activities in energy markets. One of the important subtasks of designing an effective agent for this scenario is to predict the energy use and generation of the customer agents in the marketplace. These predictions can inform pricing and tariff design questions, as well as decisions to balance power use and generation over time. Similar prediction problems are also important in real world energy markets. Here we present some initial experiments applying machine learning to predict future customer energy usage patterns in the PowerTAC simulation.