In this paper, we study the task of embodied interactive learning for object detection. Given a set of environments (and some labeling budget), our goal is to learn an object detector by having an agent select what data to obtain labels for. How should an exploration policy decide which trajectory should be labeled? One possibility is to use a trained object detector's failure cases as an external reward. However, this will require labeling millions of frames required for training RL policies, which is infeasible. Instead, we explore a self-supervised approach for training our exploration policy by introducing a notion of semantic curiosity. Our semantic curiosity policy is based on a simple observation -- the detection outputs should be consistent. Therefore, our semantic curiosity rewards trajectories with inconsistent labeling behavior and encourages the exploration policy to explore such areas. The exploration policy trained via semantic curiosity generalizes to novel scenes and helps train an object detector that outperforms baselines trained with other possible alternatives such as random exploration, prediction-error curiosity, and coverage-maximizing exploration.

Network densification and millimeter-wave technologies are key enablers to fulfill the capacity and data rate requirements of the fifth generation (5G) of mobile networks. In this context, designing low-complexity policies with local observations, yet able to adapt the user association with respect to the global network state and to the network dynamics is a challenge. In fact, the frameworks proposed in literature require continuous access to global network information and to recompute the association when the radio environment changes. With the complexity associated to such an approach, these solutions are not well suited to dense 5G networks. In this paper, we address this issue by designing a scalable and flexible algorithm for user association based on multi-agent reinforcement learning. In this approach, users act as independent agents that, based on their local observations only, learn to autonomously coordinate their actions in order to optimize the network sum-rate. Since there is no direct information exchange among the agents, we also limit the signaling overhead. Simulation results show that the proposed algorithm is able to adapt to (fast) changes of radio environment, thus providing large sum-rate gain in comparison to state-of-the-art solutions.

Can we predict how well a team of individuals will perform together? How should individuals be rewarded for their contributions to the team performance? Cooperative game theory gives us a powerful set of tools for answering these questions: the Characteristic Function (CF) and solution concepts like the Shapley Value (SV). There are two major difficulties in applying these techniques to real world problems: first, the CF is rarely given to us and needs to be learned from data. Second, the SV is combinatorial in nature. We introduce a parametric model called cooperative game abstractions (CGAs) for estimating CFs from data. CGAs are easy to learn, readily interpretable, and crucially allow linear-time computation of the SV. We provide identification results and sample complexity bounds for CGA models as well as error bounds in the estimation of the SV using CGAs. We apply our methods to study teams of artificial RL agents as well as real world teams from professional sports.

Complex black-box machine learning models are regularly used in critical decision-making domains. This has given rise to several calls for algorithmic explainability. Many explanation algorithms proposed in literature assign importance to each feature individually. However, such explanations fail to capture the joint effects of sets of features. Indeed, few works so far formally analyze \coloremph{high dimensional model explanations}. In this paper, we propose a novel high dimension model explanation method that captures the joint effect of feature subsets. We propose a new axiomatization for a generalization of the Banzhaf index; our method can also be thought of as an approximation of a black-box model by a higher-order polynomial. In other words, this work justifies the use of the generalized Banzhaf index as a model explanation by showing that it uniquely satisfies a set of natural desiderata and that it is the optimal local approximation of a black-box model. Our empirical evaluation of our measure highlights how it manages to capture desirable behavior, whereas other measures that do not satisfy our axioms behave in an unpredictable manner.

We propose Federated Generative Adversarial Network (FedGAN) for training a GAN across distributed sources of non-independent-and-identically-distributed data sources subject to communication and privacy constraints. Our algorithm uses local generators and discriminators which are periodically synced via an intermediary that averages and broadcasts the generator and discriminator parameters. We theoretically prove the convergence of FedGAN with both equal and two time-scale updates of generator and discriminator, under standard assumptions, using stochastic approximations and communication efficient stochastic gradient descents. We experiment FedGAN on toy examples (2D system, mixed Gaussian, and Swiss role), image datasets (MNIST, CIFAR-10, and CelebA), and time series datasets (household electricity consumption and electric vehicle charging sessions). We show FedGAN converges and has similar performance to general distributed GAN, while reduces communication complexity. We also show its robustness to reduced communications.

We study the problem of sequential task assignment and collision-free routing for large teams of robots in applications with inter-task precedence constraints (e.g., task $A$ and task $B$ must both be completed before task $C$ may begin). Such problems commonly occur in assembly planning for robotic manufacturing applications, in which sub-assemblies must be completed before they can be combined to form the final product. We propose a hierarchical algorithm for computing makespan-optimal solutions to the problem. The algorithm is evaluated on a set of randomly generated problem instances where robots must transport objects between stations in a "factory "grid world environment. In addition, we demonstrate in high-fidelity simulation that the output of our algorithm can be used to generate collision-free trajectories for non-holonomic differential-drive robots.

Reputation is crucial to enabling human or software agents to select among alternative providers. Although several effective reputation assessment methods exist, they typically distil reputation into a numerical representation, with no accompanying explanation of the rationale behind the assessment. Such explanations would allow users or clients to make a richer assessment of providers, and tailor selection according to their preferences and current context. In this paper, we propose an approach to explain the rationale behind assessments from quantitative reputation models, by generating arguments that are combined to form explanations. Our approach adapts, extends and combines existing approaches for explaining decisions made using multi-attribute decision models in the context of reputation. We present example argument templates, and describe how to select their parameters using explanation algorithms. Our proposal was evaluated by means of a user study, which followed an existing protocol. Our results give evidence that although explanations present a subset of the information of trust scores, they are sufficient to equally evaluate providers recommended based on their trust score. Moreover, when explanation arguments reveal implicit model information, they are less persuasive than scores.

Autonomous systems - such as cars, planes, and trains - must come with strong safety guarantees. These systems are cyber-physical, in the sense that their safety depends crucially upon the way in which their software ("cyber") components interact with their kinetic components. Cyber-physical systems (CPS) analysis tools can verify the safety of CPS by stating correctness specifications in a formal language and then verifying - via computer-checked proof - that safety-critical software components respect these specifications. Existing approaches toward formally verifying the correctness of cyber-physical systems focus primarily on constructing formal safety proofs about classical low-dimensional models of control systems. For example, the safety of an adaptive cruise control system might be established by modeling the dynamics of two cars in terms of their positions and velocities and then proving that a control policy preserves safe separation between all cars on the road for any time horizon [15]. Researchers have employed a similar approach for ensuring the correctness of proposed FAA aircraft collision avoidance protocols [12], the European Train Control System [20], and quadcopters [21]. These proofs are typically constructed and checked using a cyber-physical systems verification tool such as Flow* [4], KeYmaera X [8], or SpaceEx [6]. CPS verification tools can provide very strong safety guarantees for cyber-physical systems, but typical techniques for using these tools rely on three assumptions that break down when applying verification techniques to real autonomous systems: 1. CPS verification techniques assume that a symbolic representation of the state of the world is known a priori. For example, formal CPS models of ground robots typically assume that the system knows the positions of all relevant obstacles, at least within some error bound [16].

Optimal auctions maximize a seller's expected revenue subject to individual rationality and strategyproofness for the buyers. Myerson's seminal work in 1981 settled the case of auctioning a single item; however, subsequent decades of work have yielded little progress moving beyond a single item, leaving the design of revenue-maximizing auctions as a central open problem in the field of mechanism design. A recent thread of work in "differentiable economics" has used tools from modern deep learning to instead learn good mechanisms. We focus on the RegretNet architecture, which can represent auctions with arbitrary numbers of items and participants; it is trained to be empirically strategyproof, but the property is never exactly verified leaving potential loopholes for market participants to exploit. We propose ways to explicitly verify strategyproofness under a particular valuation profile using techniques from the neural network verification literature. Doing so requires making several modifications to the RegretNet architecture in order to represent it exactly in an integer program. We train our network and produce certificates in several settings, including settings for which the optimal strategyproof mechanism is not known.

In this work we solve the Dynamic Vehicle Routing Problem (DVRP). DVRP is a modification of the Vehicle Routing Problem, in which the clients' requests (cities) number and location might not be known at the beginning of the working day Additionally, all requests must be served during one working day by a fleet of vehicles with limited capacity. In this work we propose a Monte Carlo method (MCTree), which directly approaches the dynamic nature of arriving requests in the DVRP. The method is also hybridized (MCTree+PSO) with our previous Two-Phase Multi-swarm Particle Swarm Optimization (2MPSO) algorithm. Our method is based on two assumptions. First, that we know a bounding rectangle of the area in which the requests might appear. Second, that the initial requests' sizes and frequency of appearance are representative for the yet unknown clients' requests. In order to solve the DVRP we divide the working day into several time slices in which we solve a static problem. In our Monte Carlo approach we randomly generate the unknown clients' requests with uniform spatial distribution over the bounding rectangle and requests' sizes uniformly sampled from the already known requests' sizes. The solution proposal is constructed with the application of a clustering algorithm and a route construction algorithm. The MCTree method is tested on a well established set of benchmarks proposed by Kilby et al. and is compared with the results achieved by applying our previous 2MPSO algorithm and other literature results. The proposed MCTree approach achieves a better time to quality trade-off then plain heuristic algorithms. Moreover, a hybrid MCTree+PSO approach achieves better time to quality trade-off then 2MPSO for small optimization time limits, making the hybrid a good candidate for handling real world scale goods delivery problems.