Reinforced random forest

Abstract: This paper presents a comprehensive survey of the state-of-the-art work on fault tolerance methods proposed for cloud computing. The survey classifies fault-tolerance methods into three categories: 1) ReActive Methods (RAMs); 2) PRoactive Methods (PRMs); and 3) ReSilient Methods (RSMs). RAMs allow the system to enter into a fault status and then try to recover the system. PRMs tend to prevent the system from entering a fault status by implementing mechanisms that enable them to avoid errors before they affect the system. On the other hand, recently emerging RSMs aim to minimize the amount of time it takes for a system to recover from a fault. Machine Learning and Artificial Intelligence have played an active role in RSM domain in such a way that the recovery time is mapped to a function to be optimized (i.e., by converging the recovery time to a fraction of milliseconds). As the system learns to deal with new faults, the recovery time will become shorter. In addition, current issues and challenges in cloud fault tolerance are also discussed to identify promising areas for future research.

Abstract: Human activity recognition~(HAR) has attracted significant research interest due to its applications in health monitoring and patient rehabilitation. Recent research on HAR focuses on using smartphones due to their widespread use. However, this leads to inconvenient use, limited choice of sensors and inefficient use of resources, since smartphones are not designed for HAR. This paper presents the first HAR framework that can perform both online training and inference. The proposed framework starts with a novel technique that generates features using the fast Fourier and discrete wavelet transforms of a textile-based stretch sensor and accelerometer. Using these features, we design an artificial neural network classifier which is trained online using the policy gradient algorithm. Experiments on a low power IoT device (TI-CC2650 MCU) with nine users show 97.7% accuracy in identifying six activities and their transitions with less than 12.5 mW power consumption.

Abstract: Human activity recognition (HAR) has attracted significant research interest due to its applications in health monitoring and patient rehabilitation. Recent research on HAR focuses on using smartphones due to their widespread use. However, this leads to inconvenient use, limited choice of sensors and inefficient use of resources, since smartphones are not designed for HAR. This paper presents the first HAR framework that can perform both online training and inference. The proposed framework starts with a novel technique that generates features using the fast Fourier and discrete wavelet transforms of a textile-based stretch sensor and accelerometer data. Using these features, we design a neural network classifier which is trained online using the policy gradient algorithm. Experiments on a low power IoT device (T1-CC2650 MCU) with nine users show 97.7% accuracy in identifying six activities and their transitions with less than 12.5 mW power consumption.

Abstract: Classification using cross-datasets (where a classifier trained using annotated image set A is used to test similar images of set B due to lack of training images in B) is important for many classification problems especially in biomedical imaging. We propose a discriminative autoencoder, useful for addressing the challenge of classification using cross-datasets. Our autoencoder learns an encoder and decoder such that the distances between the representations of the same class is minimized whereas the distances between the representations of different classes are maximized. We derive a fast algorithm to solve the aforementioned problem using the Augmented Lagrangian Alternating Directions Method of Multipliers (ADMM) approach. ADMM is a faster alternative to back-propagation which is used in standard autoencoders. The proposed method outperforms state-of-the-art representation learning tools in terms of classification results in breast cancer related histopathological image set MITOS and AMIDA and some of the benchmark image datasets.

Abstract: Random forest (RF) is a supervised, non-parametric, ensemble-based machine learning method used for classification and regression task. It is easy in terms of implementation and scalable, hence attracting many researchers. Being an ensemble-based method, it considers equal weights/votes to all atomic units i.e. decision trees. However, this may not be true always for varying test cases. Hence, the correlation between decision tree and data samples are explored in the recent past to take care of such issues. In this paper, a dynamic weighing scheme is proposed between test samples and decision tree in RF. The correlation is defined in terms of similarity between the test case and the decision tree using exponential distribution. Hence, the proposed method named as Exponentially Weighted Random Forest (EWRF). The performance of the proposed method is rigorously tested over benchmark datasets from the UCI repository for both classification and regression tasks.

Abstract: Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to Adaboost (Y. Freund & R. Schapire, Machine Learning: Proceedings of the Thirteenth International conference, aaa, 148–156), but are more robust with respect to noise. Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the splitting. Internal estimates are also used to measure variable importance. These ideas are also applicable to regression.

Abstract: In an earlier paper, we introduced a new "boosting" algorithm called AdaBoost which, theoretically, can be used to significantly reduce the error of any learning algorithm that con- sistently generates classifiers whose performance is a little better than random guessing. We also introduced the related notion of a "pseudo-loss" which is a method for forcing a learning algorithm of multi-label concepts to concentrate on the labels that are hardest to discriminate. In this paper, we describe experiments we carried out to assess how well AdaBoost with and without pseudo-loss, performs on real learning problems. We performed two sets of experiments. The first set compared boosting to Breiman's "bagging" method when used to aggregate various classifiers (including decision trees and single attribute- value tests). We compared the performance of the two methods on a collection of machine-learning benchmarks. In the second set of experiments, we studied in more detail the performance of boosting using a nearest-neighbor classifier on an OCR problem.

Abstract: This paper surveys the field of reinforcement learning from a computer-science perspective. It is written to be accessible to researchers familiar with machine learning. Both the historical basis of the field and a broad selection of current work are summarized. Reinforcement learning is the problem faced by an agent that learns behavior through trial-and-error interactions with a dynamic environment. The work described here has a resemblance to work in psychology, but differs considerably in the details and in the use of the word ``reinforcement.'' The paper discusses central issues of reinforcement learning, including trading off exploration and exploitation, establishing the foundations of the field via Markov decision theory, learning from delayed reinforcement, constructing empirical models to accelerate learning, making use of generalization and hierarchy, and coping with hidden state. It concludes with a survey of some implemented systems and an assessment of the practical utility of current methods for reinforcement learning.