Now we will compile the model using optimizer as stochastic gradient descent, loss as cross-entropy and metrics to measure the performance would be accuracy. After compiling we will train the model and check the performance on validation data. We are taking a batch size of 64 and epochs to be 10.

Interpreting linear regression coefficients is common to do, because it's so easy. Training a model can be done in a few lines of code, and the results yield statistics that can be stated matter-of-factly: "each additional point on the SAT increases your chances of admission by 0.002%". Whenever you train a linear regression (or logistic regression) model with this intent, be wary: you are treading in dangerous waters. What is linear regression even doing? It multiplies each of the inputs by a value and adds them up -- as an additional degree of freedom, an'intercept' can be added.

I am predicting the selling price of the car based on various features of the cars, including the present price of the cars. I will be using Multiple Linear Regression for building The model. Let's dive under to understand the variables and use the correlation matrix to make the process easy. Now let's check if we have Outliers in our data. So Rather then removing the outliers values we would like to take log of them.

Machine learning is a pioneer subset of Artificial Intelligence, where Machines learn by itself using the available dataset. For the optimization of any machine learning model, an acceptable loss function must be selected. A Loss function characterizes how well the model performs over the training dataset. Loss functions express the discrepancy between the predictions of the model being trained and also the actual problem instances. If the deviation between predicted result and actual results is too much, then loss function would have a very high value.

If you work in computer vision, you probably know that object detection is an important topic with a lot of solutions flying around. Especially the new ones which use deep learning. In general, these solutions are separated into two types: one-stage detectors and two-stage detectors. The main difference is that two-stage detector a number of object box proposals and then perform classification and localization tasks on them, while one-stage detectors predict the object class directly from the pre-designed candidate boxes – anchors. This difference causes one-stage detectors to be faster, but two-stage detectors to have higher accuracy.

The support vector machines algorithm is a supervised machine learning algorithm that can be used for both classification and regression. In this article, we will be discussing certain parameters concerning the support vector machines and try to understand this algorithm in detail. For understanding, let us consider the SVM used for classification. The following figure shows the geometrical representation of the SVM classification. After taking a look at the above diagram you might notice that the SVM classifies the data a bit differently as compared to the other algorithms.

Graphs have become increasingly popular in modeling structures and interactions in a wide variety of problems during the last decade. Graph-based clustering and semi-supervised classification techniques have shown impressive performance. This paper proposes a graph learning framework to preserve both the local and global structure of data. Specifically, our method uses the self-expressiveness of samples to capture the global structure and adaptive neighbor approach to respect the local structure. Furthermore, most existing graph-based methods conduct clustering and semi-supervised classification on the graph learned from the original data matrix, which doesn't have explicit cluster structure, thus they might not achieve the optimal performance. By considering rank constraint, the achieved graph will have exactly $c$ connected components if there are $c$ clusters or classes. As a byproduct of this, graph learning and label inference are jointly and iteratively implemented in a principled way. Theoretically, we show that our model is equivalent to a combination of kernel k-means and k-means methods under certain condition. Extensive experiments on clustering and semi-supervised classification demonstrate that the proposed method outperforms other state-of-the-art methods.

In this thesis, we focus on video action understanding problems from an online and real-time processing point of view. We start with the conversion of the traditional offline spatiotemporal action detection pipeline into an online spatiotemporal action tube detection system. An action tube is a set of bounding connected over time, which bounds an action instance in space and time. Next, we explore the future prediction capabilities of such detection methods by extending the an existing action tube into the future by regression. Later, we seek to establish that online/causal representations can achieve similar performance to that of offline three dimensional (3D) convolutional neural networks (CNNs) on various tasks, including action recognition, temporal action segmentation and early prediction.

Modern machine learning applications should be able to address the intrinsic challenges arising over inference on massive real-world datasets, including scalability and robustness to outliers. Despite the multiple benefits of Bayesian methods (such as uncertainty-aware predictions, incorporation of experts knowledge, and hierarchical modeling), the quality of classic Bayesian inference depends critically on whether observations conform with the assumed data generating model, which is impossible to guarantee in practice. In this work, we propose a variational inference method that, in a principled way, can simultaneously scale to large datasets, and robustify the inferred posterior with respect to the existence of outliers in the observed data. Reformulating Bayes theorem via the $\beta$-divergence, we posit a robustified pseudo-Bayesian posterior as the target of inference. Moreover, relying on the recent formulations of Riemannian coresets for scalable Bayesian inference, we propose a sparse variational approximation of the robustified posterior and an efficient stochastic black-box algorithm to construct it. Overall our method allows releasing cleansed data summaries that can be applied broadly in scenarios including structured data corruption. We illustrate the applicability of our approach in diverse simulated and real datasets, and various statistical models, including Gaussian mean inference, logistic and neural linear regression, demonstrating its superiority to existing Bayesian summarization methods in the presence of outliers.

Class-Incremental Learning (CIL) aims to train a reliable model with the streaming data, which emerges unknown classes sequentially. Different from traditional closed set learning, CIL has two main challenges: 1) Novel class detection. The initial training data only contains incomplete classes, and streaming test data will accept unknown classes. Therefore, the model needs to not only accurately classify known classes, but also effectively detect unknown classes; 2) Model expansion. After the novel classes are detected, the model needs to be updated without re-training using entire previous data. However, traditional CIL methods have not fully considered these two challenges, first, they are always restricted to single novel class detection each phase and embedding confusion caused by unknown classes. Besides, they also ignore the catastrophic forgetting of known categories in model update. To this end, we propose a Class-Incremental Learning without Forgetting (CILF) framework, which aims to learn adaptive embedding for processing novel class detection and model update in a unified framework. In detail, CILF designs to regularize classification with decoupled prototype based loss, which can improve the intra-class and inter-class structure significantly, and acquire a compact embedding representation for novel class detection in result. Then, CILF employs a learnable curriculum clustering operator to estimate the number of semantic clusters via fine-tuning the learned network, in which curriculum operator can adaptively learn the embedding in self-taught form. Therefore, CILF can detect multiple novel classes and mitigate the embedding confusion problem. Last, with the labeled streaming test data, CILF can update the network with robust regularization to mitigate the catastrophic forgetting. Consequently, CILF is able to iteratively perform novel class detection and model update.