Deep Reinforcement Learning for Routing a Heterogeneous Fleet of Vehicles

Motivated by the promising advances of deep-reinforcement learning (DRL) applied to cooperative multi-agent systems we propose a model and learning procedure to solve the Capacitated Multi-Vehicle Routing Problem (CMVRP) with fixed fleet size. Our learning procedure follows a centralized-training and decentralized-execution paradigm. We empirically test our model and showed its capability for producing near-optimal solutions through cooperative actions. In large instances, our model generates better solutions than other commonly used heuristics. Additionally, our model can solve arbitrary instances of the CMVRP without requiring re-training.

A fast multi-object tracking system using an object detector ensemble

Multiple-Object Tracking (MOT) is of crucial importance for applications such as retail video analytics and video surveillance. Object detectors are often the computational bottleneck of modern MOT systems, limiting their use for real-time applications. In this paper, we address this issue by leveraging on an ensemble of detectors, each running every f frames. We measured the performance of our system in the MOT16 benchmark. The proposed model surpassed other online entries of the MOT16 challenge in speed, while maintaining an acceptable accuracy.

Learning from multivariate discrete sequential data using a restricted Boltzmann machine model

A restricted Boltzmann machine (RBM) is a generative neural-network model with many novel applications such as collaborative filtering and acoustic modeling. An RBM lacks the capacity to retain memory, making it inappropriate for dynamic data modeling as in time-series analysis. In this paper we address this issue by proposing the p-RBM model, a generalization of the regular RBM model, capable of retaining memory of p past states. We further show how to train the p-RBM model using contrastive divergence and test our model on the problem of predicting the stock market direction considering 100 stocks of the NASDAQ-100 index. Obtained results show that the p-RBM offer promising prediction potential.

Collaborative Filtering using Denoising Auto-Encoders for Market Basket Data

Recommender systems (RS) help users navigate large sets of items in the search for "interesting" ones. One approach to RS is Collaborative Filtering (CF), which is based on the idea that similar users are interested in similar items. Most model-based approaches to CF seek to train a machine-learning/data-mining model based on sparse data; the model is then used to provide recommendations. While most of the proposed approaches are effective for small-size situations, the combinatorial nature of the problem makes it impractical for medium-to-large instances. In this work we present a novel approach to CF that works by training a Denoising Auto-Encoder (DAE) on corrupted baskets, i.e., baskets from which one or more items have been removed. The DAE is then forced to learn to reconstruct the original basket given its corrupted input. Due to recent advancements in optimization and other technologies for training neural-network models (such as DAE), the proposed method results in a scalable and practical approach to CF. The contribution of this work is twofold: (1) to identify missing items in observed baskets and, thus, directly providing a CF model; and, (2) to construct a generative model of baskets which may be used, for instance, in simulation analysis or as part of a more complex analytical method.

A Probabilistic Adaptive Search System for Exploring the Face Space

Face recall is a basic human cognitive process performed routinely, e.g., when meeting someone and determining if we have met that person before. Assisting a subject during face recall by suggesting candidate faces can be challenging. One of the reasons is that the search space - the face space - is quite large and lacks structure. A commercial application of face recall is facial composite systems - such as Identikit, PhotoFIT, and CD-FIT - where a witness searches for an image of a face that resembles his memory of a particular offender. The inherent uncertainty and cost in the evaluation of the objective function, the large size and lack of structure of the search space, and the unavailability of the gradient concept makes this problem inappropriate for traditional optimization methods. In this paper we propose a novel evolutionary approach for searching the face space that can be used as a facial composite system. The approach is inspired by methods of Bayesian optimization and differs from other applications in the use of the skew-normal distribution as its acquisition function. This choice of acquisition function provides greater granularity, with regularized, conservative, and realistic results.

A Dynamic Bayesian Network Model for Inventory Level Estimation in Retail Marketing

Many retailers today employ inventory management systems based on Re-Order Point Policies, most of which rely on the assumption that all decreases in product inventory levels result from product sales. Unfortunately, it usually happens that small but random quantities of the product get lost, stolen or broken without record as time passes, e.g., as a consequence of shoplifting. This is usual for retailers handling large varieties of inexpensive products, e.g., grocery stores. In turn, over time these discrepancies lead to stock freezing problems, i.e., situations where the system believes the stock is above the re-order point but the actual stock is at zero, and so no replenishments or sales occur. Motivated by these issues, we model the interaction between sales, losses, replenishments and inventory levels as a Dynamic Bayesian Network (DBN), where the inventory levels are unobserved (i.e., hidden) variables we wish to estimate. We present an Expectation-Maximization (EM) algorithm to estimate the parameters of the sale and loss distributions, which relies on solving a one-dimensional dynamic program for the E-step and on solving two separate one-dimensional nonlinear programs for the M-step.