Noisy index coding problem over single-input single-output (SISO) additive white Gaussian noise (AWGN) broadcast channel has been studied in [6]. Further the performance of high priority receivers has been improved by using rotated PSK constellation in [7] for AWGN broadcast channel. In this work, we consider noisy index coding problem in which the source transmits symbols over a SISO Rayleigh fading broadcast channel. We show that even for transmission over Rayleigh fading channel minimum inter-set distance, as seen by the receiver plays a crucial role in its ML decoding performance unlike classical fading cases. For improving the performance of high priority receivers for a chosen index code of length N we are transmitting N bits over Rayleigh fading channel using a rotated M-PSK (M = 2N) constellation where half of the signal points have been rotated in such a way to maximize the minimum inter-set distance. Additionally, we account for the multiplicity of signal pairs along with inter-set distance as a factor to achieve better ML decoding performance for some index coding problems with rotated constellation for prioritized receivers over fading channel.

Multiple input multiple output (MIMO) scheme which employs Alamouti code with index coded PSK modulation over Rayleigh fading channel has been studied. In this work, for the noisy index coding problem we generalize the MIMO scheme for any space time block codes (STBCs) obtained from orthogonal designs in which the central server transmits symbols over Nt×NrMIMO Rician fading channel. We show that, for a chosen index code and 2N-PSK signal set at very high SNR, error performance of a receiver is decided by minimum inter-set distance seen by the receiver and for achieving its best ML decoding performance, we must choose the mappings that maximizes minimum inter-set distance. Further for improving the performance of high priority receivers for a chosen index code of length N we are transmitting N bits using a rotated M-PSK (M =2N) constellation.

Additive Manufacturing (AM) is a manufacturing paradigm that builds three-dimensional objects from a computer-aided design model by successively adding material layer by layer. AM has become very popular in the past decade due to its utility for fast prototyping such as 3D printing as well as manufacturing functional parts with complex geometries using processes such as laser metal deposition that would be difficult to create using traditional machining. As the process for creating an intricate part for an expensive metal such as Titanium is prohibitive with respect to cost, computational models are used to simulate the behavior of AM processes before the experimental run. However, as the simulations are computationally costly and time-consuming for predicting multiscale multi-physics phenomena in AM, physics-informed data-driven machine-learning systems for predicting the behavior of AM processes are immensely beneficial. Such models accelerate not only multiscale simulation tools but also empower real-time control systems using in-situ data. In this paper, we design and develop essential components of a scientific framework for developing a data-driven model-based real-time control system. Finite element methods are employed for solving time-dependent heat equations and developing the database. The proposed framework uses extremely randomized trees - an ensemble of bagged decision trees as the regression algorithm iteratively using temperatures of prior voxels and laser information as inputs to predict temperatures of subsequent voxels. The models achieve mean absolute percentage errors below 1% for predicting temperature profiles for AM processes. The code is made available for the research community at https://github.com/paularindam/ml-iter-additive.

Organic Solar Cells are a promising technology for solving the clean energy crisis in the world. However, generating candidate chemical compounds for solar cells is a time-consuming process requiring thousands of hours of laboratory analysis. For a solar cell, the most important property is the power conversion efficiency which is dependent on the highest occupied molecular orbitals (HOMO) values of the donor molecules. Recently, machine learning techniques have proved to be very useful in building predictive models for HOMO values of donor structures of Organic Photovoltaic Cells (OPVs). Since experimental datasets are limited in size, current machine learning models are trained on data derived from calculations based on density functional theory (DFT). Molecular line notations such as SMILES or InChI are popular input representations for describing the molecular structure of donor molecules. The two types of line representations encode different information, such as SMILES defines the bond types while InChi defines protonation. In this work, we present an ensemble deep neural network architecture, called SINet, which harnesses both the SMILES and InChI molecular representations to predict HOMO values and leverage the potential of transfer learning from a sizeable DFT-computed dataset- Harvard CEP to build more robust predictive models for relatively smaller HOPV datasets. Harvard CEP dataset contains molecular structures and properties for 2.3 million candidate donor structures for OPV while HOPV contains DFT-computed and experimental values of 350 and 243 molecules respectively. Our results demonstrate significant performance improvement from the use of transfer learning and leveraging both molecular representations.

Recent progress in big data and computer vision with deep learning models has gained a lot of attention. Deep learning has been performed on tasks such as image classification, object detection, image segmentation, image captioning, visual question and answering, using large collections of annotated images. This calls for more curated large image datasets with clearer descriptions, cleaner contents, and diversified usability. However, the curation and labeling of such datasets can be labor-intensive. In this paper, we present PinterNet, an algorithm for automatic curation and label generation from noisy textual descriptions, and also publish a big image dataset containing over 110K images automatically labeled with their themes. Our dataset is hierarchical in nature, it has high level category information which we refer as verticals with fine-grained thematic labels at lower level. This advocates a new type of hierarchical theme classification problem closer to human cognition and of business value. We provide benchmark performances using deep learning models based on AlexNet architecture with different pre-training schemes for this novel task and new data.

People have long sought answers to questions online, typically using either anonymous or pseudonymous forums or social network platforms that primarily use real names. Systems that allow anonymous communication afford freedom to explore identity and discuss taboo topics, but can result in negative disinhibited behavior such as cyberbullying. Identifiable communication systems allows one to reach a known audience and avoid negative disinhibition, but can constrain behavior with concerns about privacy and reputation. One persistent design issue is understanding how to leverage the benefits of anonymity without suffering its drawbacks. This paper presents a case study analysis of question asking on Facebook confession boards (FCBs), a tool popular on some college campuses. FCBs present a unique configuration in which members of an offline community (e.g., a university) anonymously submit content to a moderator who posts it to a Facebook page where others in the community can view it and respond. Response is via identifiable Facebook comments and likes. Our results show users asking about taboo and stigmatized topics with local others, and receiving relevant responses with little cyberbullying or negativity.