Epithelial cells migrate across wounds to repair injured tissue. Leader cells at the front of migrating sheets often drive this process. However, it is unclear how leaders emerge from an apparently homogeneous epithelial cell population. We characterized leaders emerging from epithelial monolayers in cell culture and found that they activated the stress sensor p53, which was sufficient to initiate leader cell behavior. p53 activated the cell cycle inhibitor p21 WAF1/CIP1 , which in turn induced leader behavior through inhibition of cyclin-dependent kinase activity. p53 also induced crowding hypersensitivity in leader cells such that, upon epithelial closure, they were eliminated by cell competition. Thus, mechanically induced p53 directs emergence of a transient population of leader cells that drive migration and ensures their clearance upon epithelial repair.

Abstract-Active learning (AL) is a promising way to efficiently build up training sets with minimal supervision. A learner deliberately queries specific instances to tune the classifier's model using as few labels as possible. The challenge for streaming is that the data distribution may evolve over time, and therefore the model must adapt. Another challenge is the sampling bias where the sampled training set does not reflect the underlying data distribution. In the presence of concept drift, sampling bias is more likely to occur as the training set needs to represent the whole evolving data. To tackle these challenges, we propose a novel bi-criteria AL (BAL) approach that relies on two selection criteria, namely, label uncertainty criterion and density-based criterion. While the first criterion selects instances that are the most uncertain in terms of class membership, the latter dynamically curbs the sampling bias by weighting the samples to reflect on the true underlying distribution. To design and implement these two criteria for learning from streams, BAL adopts a Bayesian online learning approach and combines online classification and online clustering through the use of online logistic regression and online growing Gaussian mixture models, respectively. Empirical results obtained on standard synthetic and real-world benchmarks show the high performance of the proposed BAL method compared with the state-of-the-art AL methods.

The classification of data streams is an interesting but also a challenging problem. A data stream may grow infinitely making it impractical for storage prior to processing and classification. Due to its dynamic nature, the underlying distribution of the data stream may change over time resulting in the so-called concept drift or the possible emergence and fading of classes, known as concept evolution. In addition, acquiring labels of data samples in a stream is admittedly expensive if not infeasible at all. In this paper, we propose a novel stream-based active learning algorithm (SAL) which is capable of coping with both concept drift and concept evolution by adapting the classification model to the dynamic changes in the stream. SAL is the first AL algorithm in the literature to explicitly take account of these concepts. Moreover, using SAL, only labels of samples that are expected to reduce the expected future error are queried. This process is done while tackling the problem of sampling bias so that samples that induce the change (i.e., drifting samples or samples coming from new classes) are queried. To efficiently implement SAL, the paper proposes the application of non-parametric Bayesian models allowing to cope with the lack of prior knowledge about the data stream. In particular, Dirichlet mixture models and the stick breaking process are adopted and adapted to meet the requirements of online learning. The empirical results obtained on real-world benchmarks demonstrate the superiority of SAL in terms of classification performance over the state-of-the-art methods using average and average class accuracy.

SUMMARYThis paper presents an adaptive distributed control strategy for n-serial-flexible-link manipulators. The proposed adaptive controller is used for flexible-link-manipulators: (1) to solve the tracking control problem in the joint space, and (2) to reduce vibrations of the links. The dynamical model of flexible link manipulators is reorganized to take the form of n interconnected subsystems. Each subsystem has a one-joint and one-link pair. The system parameters are deemed to be unknown. The adaptive distributed strategy controls one subsystem in each step, starting from the last one. The nth subsystem is controlled by assuming that the remaining subsystems are stable. Then, proceeding backward to the (n-1)th system, the same strategy is applied, and so on, until the first subsystem is reached. The gradient-based estimator is used to estimate the parameters of each subsystem. The control law of the ith subsystem uses its own estimated parameters and the estimated parameters of all upper level subsystems. The global stability of the error dynamics is proved using Lyapunov approach. This algorithm was implemented in real time on a two-flexible-link manipulator, and a comparison with the non-adaptive version shows the effectiveness of this approach.

Abstract_ This work presents a comparative study of two different control strategies for a flexible single-link robotic manipulator. The dynamic model of the flexible manipulator involves modeling the rotational base and the flexible link as rigid bodies using the Lagrange's method. The resulting system has one Degree-Of-Freedom (1 DOF). Two types of regulators are studied and discussed: the State-Feedback controller, and the Linear-Quadratic regulator (LQR). While the latter is obtained by resolving the Riccati equation, the state-feedback consists on poles placement. A simulation is performed on MATLAB (7.5.0)/SIMULINK (V7.0)®, and later on, experiments were achieved on the flexible beam Quanser module. Experimental results are presented and compared at the end of this paper. Keywords_ Single-link flexible manipulator, State-Feedback Controller, Linear-Quadratic Regulator (LQR).

No abstract

The dynamical model of a flexible Euler–Bernoulli link system subject to high-speed motions is developed. Nonlinear kinematics are considered to take into account the foreshortening effect of the link. A continuous model with boundary conditions is derived using Hamilton's principle and spatially discretized dynamical equations are derived using Lagrange's equation based on the expression of the kinetic and potential energies of the flexible link system. The effect of the links foreshortening on the dynamical response is shown for a sinusoidal torque input at the links hub. The simulation results showed that for small torque amplitudes the difference between linear and nonlinear kinematics is small. However, for large amplitudes, this difference is prominent.

Human activity recognition (HAR) is highly relevant to many real-world domains like safety, security, and in particular healthcare. The current machine learning technology of HAR is highly human-dependent which makes it costly and unreliable in non-stationary environment. Existing HAR algorithms assume that training data is collected and annotated by human a prior to the training phase. Furthermore, the data is assumed to exhibit the true characteristics of the underlying distribution. In this paper, we propose a new autonomous approach that consists of novel algorithms. In particular, we adopt active learning (AL) strategy to selectively query the user/resident about the label of particular activities in order to improve the model accuracy. This strategy helps overcome