In the world of Big Data, where data comes in high volume, high velocity and variety, making a data driven business decision become very crucial for business success. Machine Learning models plays a significant role with learning algorithms in turn these big data in to valuable insights.

As I am taking a course on Udemy on Deep Learning, I decided to put my knowledge to use and try to predict whether a visitor would make a purchase (generate revenue) or not. The dataset has been taken from UCI Machine Learning Repository. The first step is to import necessary libraries. Apart from the regular data science libraries including numpy, pandas and matplotlib, I import machine learning library sklearn and deep learning library keras. I will use keras to develop my Artificial Neural Network with tensorflow as the backend.

I recently was a guest speaker at the Stanford AI Salon on the topic of accessiblity in AI, which included a free-ranging discussion among assembled members of the Stanford AI Lab. There were a number of interesting questions and topics, so I thought I would share a few of my answers here. Q: What 3 things would you most like the general public to know about AI? AI is easier to use than the hype would lead you to believe. In my recent talk at the MIT Technology Review conference, I debunked several common myths that you must have a PhD, a giant data set, or expensive computational power to use AI. Most AI researchers are not working on getting computers to achieve human consciousness.

You are free to share this article under the Attribution 4.0 International license. A new algorithm enables robots to put pen to paper, writing words using stroke patterns similar to human handwriting. It's a step, the researchers say, toward robots that are able to communicate more fluently with human coworkers and collaborators. "Just by looking at a target image of a word or sketch, the robot can reproduce each stroke as one continuous action," says Atsunobu Kotani, an undergraduate student at Brown University who led the algorithm's development. "That makes it hard for people to distinguish if it was written by the robot or actually written by a human."

Deep learning computer vision startup allegro.ai is set to showcase its latest product offering, hosted at the Intel partner booth (booth #307), during the Embedded Vision Summit which will take place in Santa Clara, California on May 20-May 23, 2019. The company's platform and product suite simplify the process of developing and managing deep learning-powered perception solutions - such as for autonomous vehicles, medical imaging, drones, security, logistics and other use cases. The platform enables engineering and product managers to get the visibility and control they need, while research scientists focus their time on research and creative output. The result is meaningfully higher quality products, faster time-to-market, increased returns to scale, and materially lower costs. The company's investors include Robert Bosch Venture Capital GmbH, Samsung Catalyst Fund, Hyundai Motor Company, and other venture funds.

In many learning situations, resources at inference time are significantly more constrained than resources at training time. This paper studies a general paradigm, called Differentiable ARchitecture Compression (DARC), that combines model compression and architecture search to learn models that are resource-efficient at inference time. Given a resource-intensive base architecture, DARC utilizes the training data to learn which sub-components can be replaced by cheaper alternatives. The high-level technique can be applied to any neural architecture, and we report experiments on state-of-the-art convolutional neural networks for image classification. For a WideResNet with $97.2\%$ accuracy on CIFAR-10, we improve single-sample inference speed by $2.28\times$ and memory footprint by $5.64\times$, with no accuracy loss. For a ResNet with $79.15\%$ Top1 accuracy on ImageNet, we improve batch inference speed by $1.29\times$ and memory footprint by $3.57\times$ with $1\%$ accuracy loss. We also give theoretical Rademacher complexity bounds in simplified cases, showing how DARC avoids overfitting despite over-parameterization.

Knowledge distillation deals with the problem of training a smaller model (Student) from a high capacity source model (Teacher) so as to retain most of its performance. Existing approaches use either the training data or meta-data extracted from it in order to train the Student. However, accessing the dataset on which the Teacher has been trained may not always be feasible if the dataset is very large or it poses privacy or safety concerns (e.g., bio-metric or medical data). Hence, in this paper, we propose a novel data-free method to train the Student from the Teacher. Without even using any meta-data, we synthesize the Data Impressions from the complex Teacher model and utilize these as surrogates for the original training data samples to transfer its learning to Student via knowledge distillation. We, therefore, dub our method "Zero-Shot Knowledge Distillation" and demonstrate that our framework results in competitive generalization performance as achieved by distillation using the actual training data samples on multiple benchmark datasets.

During the last decades, we have witnessed a surge of interests of learning a low-dimensional space with discriminative information from one single view. Even though most of them can achieve satisfactory performance in some certain situations, they fail to fully consider the information from multiple views which are highly relevant but sometimes look different from each other. Besides, correlations between features from multiple views always vary greatly, which challenges multi-view subspace learning. Therefore, how to learn an appropriate subspace which can maintain valuable information from multi-view features is of vital importance but challenging. To tackle this problem, this paper proposes a novel multi-view dimension reduction method named Multi-view Locality Low-rank Embedding for Dimension Reduction (MvL2E). MvL2E makes full use of correlations between multi-view features by adopting low-rank representations. Meanwhile, it aims to maintain the correlations and construct a suitable manifold space to capture the low-dimensional embedding for multi-view features. A centroid based scheme is designed to force multiple views to learn from each other. And an iterative alternating strategy is developed to obtain the optimal solution of MvL2E. The proposed method is evaluated on 5 benchmark datasets. Comprehensive experiments show that our proposed MvL2E can achieve comparable performance with previous approaches proposed in recent literatures.

Transfer learning, in which a network is trained on one task and re-purposed on another, is often used to produce neural network classifiers when data is scarce or full-scale training is too costly. When the goal is to produce a model that is not only accurate but also adversarially robust, data scarcity and computational limitations become even more cumbersome. We consider robust transfer learning, in which we transfer not only performance but also robustness from a source model to a target domain. We start by observing that robust networks contain robust feature extractors. By training classifiers on top of these feature extractors, we produce new models that inherit the robustness of their parent networks. We then consider the case of "fine tuning" a network by retraining end-to-end in the target domain. When using lifelong learning strategies, this process preserves the robustness of the source network while achieving high accuracy. By using such strategies, it is possible to produce accurate and robust models with little data, and without the cost of adversarial training.

Automated prediction of public speaking performance enables novel systems for tutoring public speaking skills. We use the largest open repository---TED Talks---to predict the ratings provided by the online viewers. The dataset contains over 2200 talk transcripts and the associated meta information including over 5.5 million ratings from spontaneous visitors to the website. We carefully removed the bias present in the dataset (e.g., the speakers' reputations, popularity gained by publicity, etc.) by modeling the data generating process using a causal diagram. We use a word sequence based recurrent architecture and a dependency tree based recursive architecture as the neural networks for predicting the TED talk ratings. Our neural network models can predict the ratings with an average F-score of 0.77 which largely outperforms the competitive baseline method.