Abstract

This paper presents a simple and effective framework to combine various data-driven machine learning (ML) algorithms for short-range reservoir inflow forecasting, including the large-scale climate phenomenon indices addressing forecasting uncertainty. Random Forest (RF), Gradient Boosting Regressor (GBR), K-Nearest Neighbors Regressor (KNN), and Long Short-Term Memory (LSTM) were employed for predicting daily reservoir inflows considering various climate phenomenon indices (e.g., Arctic Oscillation, North Atlantic Oscillation, and Southern Oscillation Index) and hydroclimatic variables (e.g., precipitation), accounting for time-lag effects. After training the individual ML algorithm, a framework was developed to create an ensemble model using a robust weighted voting regressor (VR) method to quantify forecasting uncertainty and to improve model perfor- mances. The results of the study reveal that, for 2-day forecasts, the LSTM approach has the greatest influence on prediction accuracy, followed comparably by each model. How- ever, none of the four models seem to be noticeably superior to the VR method, regardless of the prediction lead time. The developed framework was examined on a tropical reser- voir, Bhadra reservoir Tunga-Bhadra River, located in India.

Abstract

Rivers are a significant and low-cost supply of irrigation, industrial, and residential water across the world. Due to urbanization, population growth, and climate change, many rivers are now confronting an irreparable water crisis. According to a study carried out in 2018 by the Central Pollution Control Board (CPCB) of India, there are a total of 351 polluted river stretches in India. Accessing the pollution level in these rivers is a challenge and requires the measurement of several water quality indices including turbidity, coliform levels, total phosphate, etc. These indices are then used to calculate the water quality index which gives us a fair estimate of the level of pollution in water. However, the measurement of all these indices is challenging and requires the use of complex measuring tools. In this paper, we try to solve this problem by building machine learning models that can predict the water quality index just by using a subset of the indices that are required for the calculation of the water quality index. The models trained take turbidity, pH, Dissolved Oxygen (DO), Biological Oxygen Demand (BOD), total coliform, hardness as CaCo3, fluoride, ammonia, Chemical Oxygen Demand (COD) and total suspended solids as input and make predictions for water quality using regression techniques. The models are trained specifically for Bhadra river data, but the technique can be extended for other river stretches and water bodies. We use mean squared error, mean absolute error and R2 score to evaluate our models. An artificial neural network trained with 7 dense layers outperforms other regressors in terms of the evaluation metrics used. With the availability of better datasets, the technique of using regressors for the prediction of water quality could help in the better and faster assessment of water quality which can eventually lead to quicker remedial measures

Abstract

In this paper, we address the problem of auto- matic academic curriculum generation. A curricu- lum outlines definitive topics with their sub-topics and enables teachers and students to form an over- all idea of the course outcomes and goals, and a plan of what to teach and learn to achieve those goals. Automatic curriculum generation is rel- evant in modern times with the ever increasing, rapidly changing, digitally-available academic con- tent, that is too large for manual processing by human teams. Using Wikipedia as an external knowledge-base, along with a pipeline of standard components, we show that it is possible to generate human-interpretable 2-level topic hierarchies. We show that our approach works on publicly available textbooks, by first removing their title-structure, and then automatically regenerating a 2-level title structure that is on-par.

Abstract

1 Introduction Recent studies show that sexually-objectified women are not only treated as sex objects but also visually processed similar to objects (local processing)[1,2]. However, Sexual objectification(SO) research has predominantly studied western society's perceptions, with few empirical studies examining other cultures. In this study, we explore SO of women in an Asian-Indian cultural/social setting using a novel design and stimuli representative of the culture. We conduct an eye tracking study that uses Navon local/global and body inversion[3] paradigms to investigate the gaze behavior and underlying visual processing when looking at non-sexualized male and female images. 2 Methods The study sample consists of Ninety-four (24 women, 70 men, age=18-27, mean=21, stdev=±2.1) undergraduate and graduate students. There are 20 trial blocks. Each trial block consists of a male/female image in upright/inverted orientation, off-centre to the screen for 7 seconds followed by a plus sign for 1 second, followed by a Navon letter identification task. The priming condition and Navon task are randomised to minimise confounding effects. The Navon image is a large letter made of small letters. The participants press 'B' if they identify either one of the target letters ('H'/‘O') either as a contour or a feature (the letter itself) otherwise press ‘N’. The classification of local/global recognition is implicit as the participants press the same key ‘B’ for both local and global images. We selected black and white male images(MI) and female images(FI) wearing casual attires like jeans/pant and top/shirt. The experiment is designed using Tobii Studio. Tobiix120 eye tracker is used to capture the gaze. We mark face, chest, hip as our Areas of Interest(AoIs) and calculate the fixation duration(FD), revisit frequency(RF), first fixation duration(FFD) for the AoIs. We define sexual body(SB) parts as sum of hip and chest AoIs. For gaze analysis, 5 participants data with poor sampling rate (below 40%) are excluded. We use Tukey Fence outlier detection method to remove outliers and removed 10 participants data for first fixation duration analysis. We measure Response Time(RT) as the time taken to press the key from the onset of the Navon image presentation. For RT analysis, we consider only the data points with at least one correct response in each priming condition, reducing our data to 59 participants. The alpha level for p-value is set to 0.05 in all tests. 3 Results There is a main effect of target (priming stimulus) gender and orientation on the FD and RF. • FD: Face FD of upright > inverted and SB parts FD of upright < inverted for both MI and FI. Face FD of FI < MI and SB parts FD of FI > MI in both upright and inverted orientation. Face FD > SB parts FD for all target images. • RF: Face RF of upright > inverted for both MI and FI. SB parts RF of inverted > upright for FI. Importantly, SB parts RF of FI > MI in upright and inverted orien-tation. Face RF < SB parts RF for all target images. • FFD: There is significant effect of orientation for FI but not for MI. Face FFD of upright FI > inverted FI and SB parts FFD of inverted FI > upright FI. • RT: Global recognition is faster in female priming for both orientations. Local recognition is faster in male priming, but the significance is found only for inverted image. The participant gender effect on RT is not significant. There is no significant correlation between local/global RT and FD. 4 Discussion Objectification theory states that women have a greater propensity to be sexually objectified than men do. This is evident from our data, female sexual body parts received greater attention than male sexual body parts and vice versa for face. However the visual attention is still greatest on the face for both male and female images, in line with [4,5,6,7,8]. As for global/local perception, non-sexualized males are processed similar to objects more than females. This is in contrast to[3,1,6] but find support in a study on Chinese population [9]. We can attribute our findings to the choice of the stimuli being personalised as against sexualized. There is a strong inversion effect on gaze, both gender images are more objectified in inverted than upright position but this doesn’t affect the underlying cognitive processing implying that the difference in the visual processing between objectified and non-objectified is more quantitative than qualitative. References 3) Bernard, P., Gervais, S. J., & Klein, O. (2018). Objectifying objectification: When and why people are cognitively reduced to their parts akin to objects. European Review of So-cial Psychology, 29(1), 82–121. 4) Bernard, P., Rizzo, T., Hoonhorst, I., Deliens, G., Gervais, S., B, Bayard, C., Deltenre, P., Colin, C., & Klein, O. (2018). The Neural Correlates of Cognitive Objectification. So-cial Psychological and Personality Science, 9, 550–559. 2) Bernard, P., Gervais, S. J., Allen, J., Campomizzi, S., & Klein, O. (2012). Integrating Sexual Objectification with Object Versus Person Recognition: The Sexualized-Body-Inversion Hypothesis. Psychological Science, 23(5), 469–471. 1) Bareket, O., Shnabel, N., Abeles, D., Gervais, S., & Yuval-Greenberg, S. (2019). Evi-dence for an Association between Men’s Spontaneous Objectifying Gazing Behavior and their Endorsement of Objectifying Attitudes toward Women. Sex Roles, 81(3–4), 245–256. 5) Cogoni, C., Carnaghi, A., Mitrovic, A., Leder, H., Fantoni, C., & Silani, G. (2018). Understanding the mechanisms behind the sexualized-body inversion hypothesis: The role of asymmetry and attention biases. PLOS ONE

Abstract

Material science enables us to develop a wide array of materials capable enough to address worldly problems. Velostat is a carbon-impregnated polythene material used to address many problems including gait-related problems. This paper presents a novel approach to analyze a gait-cycle parameter called ‘Stance-phase.’ We ideated, designed, and implemented a low-cost customizable Velostat based sensing mat to calculate the stance phase. We validated our VelGmat using sensor density, sensitivity, process time, and hardware performance metrics. Our validation results helped us improve the effectiveness of gait analysis. We conducted a comparative study between our VelGmat and commercially available mat to understand the contrast in stance phase values. We observed 95% accuracy in stance phase values compared with commercially available mat.

Abstract

Increasingly, pre-trained convolutional neural networks (CNNs) are being deployed for inference in various computer vision applications, both on the server-side in the data centers and at the edge. CNN inference is a very compute-intensive task. It is a challenge to meet performance metrics such as latency and throughput while optimizing power. Special-purpose ASICs and FPGAs are suitable candidates to meet these power and performance budgets simultaneously. Rapidly evolving CNN architectures involve novel convolution operations such as point convolutions, depth separable convolutions, and so on. This leads to substantial variation in the computational structure across CNNs and layers within a CNN. Because of this, FPGA reconfigurability provides an attractive tradeoff compared to ASICs. FPGA-based hardware designers address the structural variability issue by generating a network-specific accelerator for a single network or a class of networks. However, homogeneous accelerators are network agnostic and often sacrifice throughput and FPGA LUTs for flexibility. In this article, we propose an FPGA overlay for efficient processing of CNNs that can be scaled based on the available compute and memory resources of the FPGA. The overlay is configured on the fly through control words sent by the host on a per-layer basis. Unlike current overlays, our architecture exploits all forms of parallelism inside a convolution operation. A constraint system is employed at the host end to find out the per-layer configuration of the overlay that uses all forms of parallelism in the processing of the layer, resulting in the highest throughput for that layer. We studied the effectiveness of our overlay by using it to process AlexNet, VGG16, YOLO, MobileNet, and ResNet-50 CNNs targeting a Virtex7 and a bigger Ultrascale+VU9P FPGAs. The chosen CNNs have a mix of different types of convolution layers and filter sizes, presenting a good variation in model size and structure. Our accelerator reported a maximum throughput of 1,200 GOps/second on the Virtex7, an improvement of 1.2× to 5× over the recent designs. Also, the reported performance density, measured in giga operations per second per KLUT, is 1.3× to 4× improvement over existing works. Similar speed-up and performance density is also observed for the Ultrascale+VU9P FPGA.

Abstract

This paper presents a fast and efficient optimization engine with multi-directional, multi-objective algorithms based on a robust transistor sizing approach to improve digital circuit performance. However, such optimization processes are highly simulator-dependent and computationally expensive tasks. There-fore, we propose developing machine learning-based reliable models considering process and operating variations to speed up the optimization procedure by running them on developed Residual Neural Network (ResNN) models instead of running expensive circuit simulations. Results on 22nm Metal Gate High-K digital cells show a reduction in delay and leakage up to 36.7% and 18.8 %, respectively improving computational efficiency by several orders.

Abstract

Quantum sensing applications such as nitrogen vacancy based magnetometry require phase locked loops (PLL), which can synthesize microwave frequencies in a narrow band near 2.87 GHz. Moreover, it is also required that the PLLs should have low noise and low jitter for high stability and fast settling time. These requirements seek low phase noise voltage controlled oscillator (VCO) with small variation in its gain (KVCO) within the desired tuning range. In this paper, we present a technique for designing a low phase noise VCO with low KVCO and small KVCO variation. To validate the proposed technique, an LC VCO has been designed and implemented in 0.18 µm CMOS process and post layout simulation results are presented. The simulation results show that the proposed LC VCO achieves KVCO variation of ±3.82% in the frequency range of 2.75 - 2.94 GHz and exhibits a phase noise of -118.64 dBc/Hz at an offset of 1 MHz, while consuming 9 mW of power from a 1.8 V supply.

Abstract

Precision farming has a positive potential in the agricultural industry regarding water conservation, increased productivity, better development of rural areas, and increased income. Blockchain technology is a better alternative for storing and sharing farm data as it is reliable, transparent, immutable, and decentralized. Remote monitoring of an agricultural field requires security systems to ensure that any sensitive information is exchanged only among authenticated entities in the network. To this end, we design an efficient blockchain-enabled authenticated key agreement scheme for mobile vehicles-assisted precision agricultural Internet of Things (IoT) networks called AgroMobiBlock . The limited existing work on authentication in agricultural networks shows passive usage of blockchains with very high costs. AgroMobiBlock proposes a novel idea using the elliptic curve operations on an active hybrid blockchain over mobile farming vehicles with low computation and communication costs. Formal and informal security analysis along with the formal security verification using the Automated Validation of Internet Security Protocols and Applications (AVISPA) software tool have shown the robustness of AgroMobiBlock against man-in-the-middle, impersonation, replay, physical capture, and ephemeral secret leakage attacks among other potential attacks. The blockchain-based simulation on large-scale nodes shows the computational time for an increase in the network and block sizes. Moreover, the real-time testbed experiments have been performed to show the practical usefulness of the proposed scheme.

Abstract

Coral Reefs are vital ecosystems as they support a vast population of aquatic life and help in maintaining biodiversity. Currently, these ecosystems are under severe threat due to rising ocean temperatures and contaminated surroundings. Hence, Coral Reefs need to be monitored regularly to study their health and protect them in a better way. Remote detection and assessment of Coral Reefs using satellite data would significantly enhance the reach and ensure continuous Coral Reef monitoring. The spectral signature of the Coral Reefs captured by the satellite sensor could be a great asset in this process. Still, these signatures are heavily dopped with several unwanted noise signals coming from various surface and atmospheric features. It is essential to remove these noise signals from the satellite data. This paper studied the effectiveness of different available atmospheric correction methods such as QUAC, 6S, and FLAASH Atmospheric correction tools (ACT) in removing atmospheric noise and studied the impact of sun glint and water column correction on improving remote detection of Coral Reefs. As a case study, these processes were applied on Landsat ETM+ to study the abundant Coral habitat near the North Reef Island at the Andaman Islands. The results indicated that FLAASH is the best atmospheric correction tool among the selected algorithms as FLAASH corrected image facilitated better identification of Coral Reefs. Hence this image was further enhanced using the Hedley sun glint and Lyzenga water column correction methods which improved Coral Reef detectability significantly. The detection of Corals from the image was done using ISO-DATA unsupervised classification to evaluate the role of each process in enhancing the Coral detection. Visual inspection was used to assess the classifier accuracy towards Coral reef detection. The results showed that a combination of FLAASH, sun glint removal, and water column correction significantly improved Coral Reef identification from the remote sensing data.