It is observed that the Engineering students are coming in with weakness in problem solving skills. For example, in Electric Circuits I course, they are able to apply principles of Ohm’s Law, Kirchhoff’s laws, mesh and nodal analysis techniques. However, they can obtain the solution of resulting simultaneous equations correctly 80% of the time. They are making mistakes in solving two or three variable algebraic simultaneous equations. Therefore, 90% of mistakes are in calculation of unknown variables. This is also true in Electric Circuits II course where they have difficulties in applying calculus and differential equations. This type of weakness exists in all branches of engineering. One way is to actively engage students with some intervention. We have designed an intervention mechanism and have been experimenting for the last two years. This paper describes intervention experiment and results from tracking of their cumulative GPAs for three-year period.

The nation is reaching at a critical stage due to shortage of Science Technology Engineering Mathematics (STEM) related workers. The shortage is felt across all government agencies and private sector. High tech industry is bringing foreign workers via H1B program to fill their needs. This is creating a challenge within the country and becoming a national security issue. One way to keep students in the STEM field is to increase recruitment, retention, and graduation rate. This paper is focused on providing one experiment to achieve above goals.

It has been reported that incoming students are weak in Mathematics and thereby having great difficulty in engineering courses. As a result, they are dropping out of engineering programs resulting in low retention and graduation rates. It is desired to reinforce mathematical concepts in gateway courses. One way is to engage them via additional informal instruction sessions, peer mentoring research projects. Our experiment with five activities of offering Saturday Academy, peer mentoring, professional lecture series, freshman seminar and summer research projects has been running for two semesters. Results are preliminary and future looks optimistic.

We consider the problem of training User Verification (UV) models in federated setup, where each user has access to the data of only one class and user embeddings cannot be shared with the server or other users. To address this problem, we propose Federated User Verification (FedUV), a framework in which users jointly learn a set of vectors and maximize the correlation of their instance embeddings with a secret linear combination of those vectors. We show that choosing the linear combinations from the codewords of an error-correcting code allows users to collaboratively train the model without revealing their embedding vectors. We present the experimental results for user verification with voice, face, and handwriting data and show that FedUV is on par with existing approaches, while not sharing the embeddings with other users or the server.

Machine learning tasks typically involve minimizing a loss function that measures the distance of the model output and the ground-truth. In some applications, in addition to the usual loss function, the output must also satisfy certain requirements for further processing. We call such requirements model conditioning. We investigate cases where the conditioner is not differentiable or cannot be expressed in closed form and, hence, cannot be directly included in the loss function of the machine learning model. We propose to replace the conditioner with a learned dummy model which is applied on the output of the main model. The entire model, composed of the main and dummy models, is trained end-to-end. Throughout training, the dummy model learns to approximate the conditioner and, thus, forces the main model to generate outputs that satisfy the specified requirements. We demonstrate our approach on a use-case of demand charge-aware electricity load forecasting. We show that jointly minimizing the error in forecast load and its demand charge threshold results in significant improvement to existing load forecast methods.

Accurate modeling of battery capacity degradation is an important component for both battery manufacturers and energy management systems. In this paper, we develop a battery degradation model using deep learning algorithms. The model is trained with the real data collected from battery storage solutions installed and operated for behind-the-meter customers. In the dataset, battery operation data are recorded at a small scale (five minutes) and battery capacity is measured at every six months. In order to improve the training performance, we apply two preprocessing techniques, namely subsampling and feature extraction on operation data, and also interpolating between capacity measurements at times for which battery operation features are available. We integrate both cyclic and calendar aging processes in a unified framework by extracting the corresponding features from operation data. The proposed model uses LSTM units followed by a fully-connected network to process weekly battery operation features and predicts the capacity degradation. The experimental results show that our method can accurately predict the capacity fading and significantly outperforms baseline models including persistence and autoregressive (AR) models.

Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance on a variety of computer vision tasks, particularly visual classification problems, where new algorithms reported to achieve or even surpass the human performance. In this paper, we examine whether CNNs are capable of learning the semantics of training data. To this end, we evaluate CNNs on negative images, since they share the same structure and semantics as regular images and humans can classify them correctly. Our experimental results indicate that when training on regular images and testing on negative images, the model accuracy is significantly lower than when it is tested on regular images. This leads us to the conjecture that current training methods do not effectively train models to generalize the concepts. We then introduce the notion of semantic adversarial examples - transformed inputs that semantically represent the same objects, but the model does not classify them correctly - and present negative images as one class of such inputs.

Google has recently introduced the Cloud Vision API for image analysis. According to the demonstration website, the API "quickly classifies images into thousands of categories, detects individual objects and faces within images, and finds and reads printed words contained within images." It can be also used to "detect different types of inappropriate content from adult to violent content." In this paper, we evaluate the robustness of Google Cloud Vision API to input perturbation. In particular, we show that by adding sufficient noise to the image, the API generates completely different outputs for the noisy image, while a human observer would perceive its original content. We show that the attack is consistently successful, by performing extensive experiments on different image types, including natural images, images containing faces and images with texts. For instance, using images from ImageNet dataset, we found that adding an average of 14.25% impulse noise is enough to deceive the API. Our findings indicate the vulnerability of the API in adversarial environments. For example, an adversary can bypass an image filtering system by adding noise to inappropriate images. We then show that when a noise filter is applied on input images, the API generates mostly the same outputs for restored images as for original images. This observation suggests that cloud vision API can readily benefit from noise filtering, without the need for updating image analysis algorithms.

We consider the setting where a collection of time series, modeled as random processes, evolve in a causal manner, and one is interested in learning the graph governing the relationships of these processes. A special case of wide interest and applicability is the setting where the noise is Gaussian and relationships are Markov and linear. We study this setting with two additional features: firstly, each random process has a hidden (latent) state, which we use to model the internal memory possessed by the variables (similar to hidden Markov models). Secondly, each variable can depend on its latent memory state through a random lag (rather than a fixed lag), thus modeling memory recall with differing lags at distinct times. Under this setting, we develop an estimator and prove that under a genericity assumption, the parameters of the model can be learned consistently. We also propose a practical adaption of this estimator, which demonstrates significant performance gains in both synthetic and real-world datasets.

In this letter, we propose a method for real-time high density impulse noise suppression from images. In our method, we first apply an impulse detector to identify the corrupted pixels and then employ an innovative weighted-average filter to restore them. The filter takes the nearest neighboring interpolated image as the initial image and computes the weights according to the relative positions of the corrupted and uncorrupted pixels. Experimental results show that the proposed method outperforms the best existing methods in both PSNR measure and visual quality and is quite suitable for real-time applications.

We present a methodology and a tool for objective characterization of voice service quality in W-CDMA networks. Specifically, the speech quality of the UMTS adaptive multi-rate speech codec is investigated and relevant results are presented. The results show that the speech quality is not affected appreciably by transmission errors when the FER is kept below 1%. However, the codec can be operated at up to FER=7% in the highest codec mode while keeping a voice quality as good as or better than lower codec rates at lower FER.