Say you've got a dataset where there exist relationships between individual samples, and your goal is to identify groups of related samples within the dataset. Clustering, which is part of the class of unsupervised machine learning algorithms, is then the way to go. But what clustering algorithm to apply when you do not really know the number of clusters? Enter Affinity Propagation, a gossip-style algorithm which derives the number of clusters by mimicing social group formation by passing messages about the popularity of individual samples as to whether they're part of a certain group, or even if they are the leader of one. This algorithm, which can estimate the number of clusters/groups in your dataset itself, is the topic of today's blog post.

While doing ML algorithms like linear Regression, logistic regression, etc. the algorithm uses one or more independent variables for predicting the dependent variable. The data set should be balanced to avoid predicting the results incorrectly. But it is not necessary that the data set is balanced all the time. In some cases like fraudulent data, cancer patient records, etc the data set may be imbalanced. Today we will discuss about how to clean an Imbalanced data set.

Machine learning is predicated on the concept of generalization: a model achieving low error on a sufficiently large training set should also perform well on novel samples from the same distribution. We show that both data whitening and second order optimization can harm or entirely prevent generalization. In general, model training harnesses information contained in the sample-sample second moment matrix of a dataset. For a general class of models, namely models with a fully connected first layer, we prove that the information contained in this matrix is the only information which can be used to generalize. Models trained using whitened data, or with certain second order optimization schemes, have less access to this information; in the high dimensional regime they have no access at all, producing models that generalize poorly or not at all. We experimentally verify these predictions for several architectures, and further demonstrate that generalization continues to be harmed even when theoretical requirements are relaxed. However, we also show experimentally that regularized second order optimization can provide a practical tradeoff, where training is still accelerated but less information is lost, and generalization can in some circumstances even improve.

Abstract-- In highway scenarios, an alert human driver will typically anticipate early cutin and cutout maneuvers of surrounding vehicles using only visual cues. Different sizes of the regions around the vehicles are analyzed, evaluating the importance of the interaction between vehicles and the context information in the performance. I. INTRODUCTION One of the closest and most plausible scenarios in the To deal with lane-change prediction of surrounding vehicles, adoption of the autonomous vehicles is autonomous navigation in this paper we pose the problem as an action at SAE L3 (chauffeur) or L4 (autopilot) on highways, recognition problem using visual information from cameras. The most advanced The idea behind our proposal is to use the same source of information automation systems to date are the Highway Chauffeur (visual cues) and the same type of approach (action (HC) and the Highway Autopilot (HA), which includes the recognition) that drivers use to anticipate these maneuvers. HC is mostly considered as L3 and HA as L4[1].

As an economical and healthy mode of shared transportation, Bike Sharing System (BSS) develops quickly in many big cities. An accurate prediction method can help BSS schedule resources in advance to meet the demands of users, and definitely improve operating efficiencies of it. However, most of the existing methods for similar tasks just utilize spatial or temporal information independently. Though there are some methods consider both, they only focus on demand prediction in a single location or between location pairs. In this paper, we propose a novel deep learning method called Spatial-Temporal Dynamic Interval Network (STDI-Net). The method predicts the number of renting and returning orders of multiple connected stations in the near future by modeling joint spatial-temporal information. Furthermore, we embed an additional module that generates dynamical learnable mappings for different time intervals, to include the factor that different time intervals have a strong influence on demand prediction in BSS. Extensive experiments are conducted on the NYC Bike dataset, the results demonstrate the superiority of our method over existing methods.

There are situations where data relevant to a machine learning problem are distributed among multiple locations that cannot share the data due to regulatory, competitiveness, or privacy reasons. For example, data present in users' cellphones, manufacturing data of companies in a given industrial sector, or medical records located at different hospitals. Moreover, participating sites often have different data distributions and computational capabilities. Federated Learning provides an approach to learn a joint model over all the available data in these environments. In this paper, we introduce a novel distributed validation weighting scheme (DVW), which evaluates the performance of a learner in the federation against a distributed validation set. Each learner reserves a small portion (e.g., 5%) of its local training examples as a validation dataset and allows other learners models to be evaluated against it. We empirically show that DVW results in better performance compared to established methods, such as FedAvg, both under synchronous and asynchronous communication protocols in data and computationally heterogeneous environments.

Spatiotemporal action recognition is the task of locating and classifying actions in videos. Our project applies this task to analyzing video footage of restaurant workers preparing food, for which potential applications include automated checkout and inventory management. Such videos are quite different from the standardized datasets that researchers are used to, as they involve small objects, rapid actions, and notoriously unbalanced data classes. We explore two approaches - one involving the familiar object detector "You Only Look Once" (YOLO), and another applying a recently proposed analogue for action recognition, "You Only Watch Once" (YOWO). In the first, we design and implement a novel, recurrent modification of YOLO using convolutional LSTMs and explore the various subtleties in the training of such a network. In the second, we study the ability of YOWO's three-dimensional convolutions to capture the spatiotemporal features of our unique dataset, which was generously lent by CMU-based startup Agot.

As market wearables are becoming popular with a range of services, including making financial transactions, accessing cars, etc. that they provide based on various private information of a user, security of this information is becoming very important. However, users are often flooded with PINs and passwords in this internet of things (IoT) world. Additionally, hard-biometric, such as facial or finger recognition, based authentications are not adaptable for market wearables due to their limited sensing and computation capabilities. Therefore, it is a time demand to develop a burden-free implicit authentication mechanism for wearables using the less-informative soft-biometric data that are easily obtainable from the market wearables. In this work, we present a context-dependent soft-biometric-based wearable authentication system utilizing the heart rate, gait, and breathing audio signals. From our detailed analysis, we find that a binary support vector machine (SVM) with radial basis function (RBF) kernel can achieve an average accuracy of $0.94 \pm 0.07$, $F_1$ score of $0.93 \pm 0.08$, an equal error rate (EER) of about $0.06$ at a lower confidence threshold of 0.52, which shows the promise of this work.

The ocean is filled with microscopic microalgae called phytoplankton, which together are responsible for as much photosynthesis as all plants on land combined. Our ability to predict their response to the warming ocean relies on understanding how the dynamics of phytoplankton populations is influenced by changes in environmental conditions. One powerful technique to study the dynamics of phytoplankton is flow cytometry, which measures the optical properties of thousands of individual cells per second. Today, oceanographers are able to collect flow cytometry data in real-time onboard a moving ship, providing them with fine-scale resolution of the distribution of phytoplankton across thousands of kilometers. One of the current challenges is to understand how these small and large scale variations relate to environmental conditions, such as nutrient availability, temperature, light and ocean currents. In this paper, we propose a novel sparse mixture of multivariate regressions model to estimate the time-varying phytoplankton subpopulations while simultaneously identifying the specific environmental covariates that are predictive of the observed changes to these subpopulations. We demonstrate the usefulness and interpretability of the approach using both synthetic data and real observations collected on an oceanographic cruise conducted in the north-east Pacific in the spring of 2017.

We study the multiple manifold problem, a binary classification task modeled on applications in machine vision, in which a deep fully-connected neural network is trained to separate two low-dimensional submanifolds of the unit sphere. We provide an analysis of the one-dimensional case, proving for a simple manifold configuration that when the network depth $L$ is large relative to certain geometric and statistical properties of the data, the network width $n$ grows as a sufficiently large polynomial in $L$, and the number of i.i.d. samples from the manifolds is polynomial in $L$, randomly-initialized gradient descent rapidly learns to classify the two manifolds perfectly with high probability. Our analysis demonstrates concrete benefits of depth and width in the context of a practically-motivated model problem: the depth acts as a fitting resource, with larger depths corresponding to smoother networks that can more readily separate the class manifolds, and the width acts as a statistical resource, enabling concentration of the randomly-initialized network and its gradients. The argument centers around the neural tangent kernel and its role in the nonasymptotic analysis of training overparameterized neural networks; to this literature, we contribute essentially optimal rates of concentration for the neural tangent kernel of deep fully-connected networks, requiring width $n \gtrsim L\,\mathrm{poly}(d_0)$ to achieve uniform concentration of the initial kernel over a $d_0$-dimensional submanifold of the unit sphere $\mathbb{S}^{n_0-1}$, and a nonasymptotic framework for establishing generalization of networks trained in the NTK regime with structured data. The proof makes heavy use of martingale concentration to optimally treat statistical dependencies across layers of the initial random network. This approach should be of use in establishing similar results for other network architectures.