Model-assisted Learning-based Framework for Sensor Fault-Tolerant Building HVAC Control

As people spend up to 87% of their time indoors, intelligent Heating, Ventilation, and Air Conditioning (HVAC) systems in buildings are essential for maintaining occupant comfort and reducing energy consumption. Those HVAC systems in modern smart buildings rely on real-time sensor readings, which in practice often suffer from various faults and could also be vulnerable to malicious attacks. Such faulty sensor inputs may lead to the violation of indoor environment requirements (e.g., temperature, humidity, etc.) and the increase of energy consumption. While many model-based approaches have been proposed in the literature for building HVAC control, it is costly to develop accurate physical models for ensuring their performance and even more challenging to address the impact of sensor faults. In this work, we present a novel learning-based framework for sensor fault-tolerant HVAC control, which includes three deep learning based components for 1) generating temperature proposals with the consideration of possible sensor faults, 2) selecting one of the proposals based on the assessment of their accuracy, and 3) applying reinforcement learning with the selected temperature proposal. Moreover, to address the challenge of training data insufficiency in building-related tasks, we propose a model-assisted learning method leveraging an abstract model of building physical dynamics. Through extensive numerical experiments, we demonstrate that the proposed fault-tolerant HVAC control framework can significantly reduce building temperature violations under a variety of sensor fault patterns while maintaining energy efficiency.

Span-based discontinuous constituency parsing: a family of exact chart-based algorithms with time complexities from O(n^6) down to O(n^3)

We introduce a novel chart-based algorithm for span-based parsing of discontinuous constituency trees of block degree two, including ill-nested structures. In particular, we show that we can build variants of our parser with smaller search spaces and time complexities ranging from $\mathcal O(n^6)$ down to $\mathcal O(n^3)$. The cubic time variant covers 98\% of constituents observed in linguistic treebanks while having the same complexity as continuous constituency parsers. We evaluate our approach on German and English treebanks (Negra, Tiger and Discontinuous PTB) and report state-of-the-art results in the fully supervised setting. We also experiment with pre-trained word embeddings and \bert{}-based neural networks.

NB-IoT via LEO satellites: An efficient resource allocation strategy for uplink data transmission

In this paper, we focus on the use of Low-Eart Orbit (LEO) satellites providing the Narrowband Internet of Things (NB-IoT) connectivity to the on-ground user equipment (UEs). Conventional resource allocation algorithms for the NBIoT systems are particularly designed for terrestrial infrastructures, where devices are under the coverage of a specific base station and the whole system varies very slowly in time. The existing methods in the literature cannot be applied over LEO satellite-based NB-IoT systems for several reasons. First, with the movement of the LEO satellite, the corresponding channel parameters for each user will quickly change over time. Delaying the scheduling of a certain user would result in a resource allocation based on outdated parameters. Second, the differential Doppler shift, which is a typical impairment in communications over LEO, directly depends on the relative distance among users. Scheduling at the same radio frame users that overcome a certain distance would violate the differential Doppler limit supported by the NB-IoT standard. Third, the propagation delay over a LEO satellite channel is around 4-16 times higher compared to a terrestrial system, imposing the need for message exchange minimization between the users and the base station. In this work, we propose a novel uplink resource allocation strategy that jointly incorporates the new design considerations previously mentioned together with the distinct channel conditions, satellite coverage times and data demands of various users on Earth. The novel methodology proposed in this paper can act as a framework for future works in the field.

Multi-functional microwave photonic radar system for simultaneous distance and velocity measurement and high-resolution microwave imaging

A photonic-assisted multi-functional radar system for simultaneous distance and velocity measurement and high-resolution microwave imaging is proposed and experimentally demonstrated by using a composite transmitted microwave signal of a single-chirped linearly frequency-modulated (LFM) signal and a single-tone microwave signal. In the system, the transmitted signal is generated via photonic frequency up-conversion based on a single integrated dual-polarization dual-parallel Mach-Zehnder modulator (DPol-DPMZM), whereas the echo signals scattered from the target are de-chirped to two low-frequency signals using a microwave photonic frequency mixer. By using the two low-frequency de-chirped signals, the real-time distance and radial velocity of the moving target can be measured accurately according to the round-trip time of the echo signal and its Doppler frequency shift. Compared with the previous reported distance and velocity measurement methods, where two LFM signals with opposite chirps are used, these parameters can be obtained using only a single-chirped LFM signal and a single-tone microwave signal. Meanwhile, high-resolution inverse synthetic aperture radar (ISAR) imaging can also be realized using ISAR imaging algorithms. An experiment is performed to verify the proposed multi-functional microwave photonic radar system. An up-chirped LFM signal from 8.5 to 12.5 GHz and an 8.0 GHz single-tone microwave signal are used as the transmitted signal. The results show that the absolute measurement errors of distance and radial velocity are less than 5.9 cm and 2.8 cm/s, respectively. ISAR imaging results are also demonstrated, which proves the high-resolution and real-time ISAR imaging ability of the proposed system.

Efficient training for future video generation based on hierarchical disentangled representation of latent variables

Generating videos predicting the future of a given sequence has been an area of active research in recent years. However, an essential problem remains unsolved: most of the methods require large computational cost and memory usage for training. In this paper, we propose a novel method for generating future prediction videos with less memory usage than the conventional methods. This is a critical stepping stone in the path towards generating videos with high image quality, similar to that of generated images in the latest works in the field of image generation. We achieve high-efficiency by training our method in two stages: (1) image reconstruction to encode video frames into latent variables, and (2) latent variable prediction to generate the future sequence. Our method achieves an efficient compression of video into low-dimensional latent variables by decomposing each frame according to its hierarchical structure. That is, we consider that video can be separated into background and foreground objects, and that each object holds time-varying and time-independent information independently. Our experiments show that the proposed method can efficiently generate future prediction videos, even for complex datasets that cannot be handled by previous methods.

Hyperparameter Optimization: Foundations, Algorithms, Best Practices and Open Challenges

Most machine learning algorithms are configured by one or several hyperparameters that must be carefully chosen and often considerably impact performance. To avoid a time consuming and unreproducible manual trial-and-error process to find well-performing hyperparameter configurations, various automatic hyperparameter optimization (HPO) methods, e.g., based on resampling error estimation for supervised machine learning, can be employed. After introducing HPO from a general perspective, this paper reviews important HPO methods such as grid or random search, evolutionary algorithms, Bayesian optimization, Hyperband and racing. It gives practical recommendations regarding important choices to be made when conducting HPO, including the HPO algorithms themselves, performance evaluation, how to combine HPO with ML pipelines, runtime improvements, and parallelization.

Impact of Data Normalization on Deep Neural Network for Time Series Forecasting

For the last few years it has been observed that the Deep Neural Networks (DNNs) has achieved an excellent success in image classification, speech recognition. But DNNs are suffer great deal of challenges for time series forecasting because most of the time series data are nonlinear in nature and highly dynamic in behaviour. The time series forecasting has a great impact on our socio-economic environment. Hence, to deal with these challenges its need to be redefined the DNN model and keeping this in mind, data pre-processing, network architecture and network parameters are need to be consider before feeding the data into DNN models. Data normalization is the basic data pre-processing technique form which learning is to be done. The effectiveness of time series forecasting is heavily depend on the data normalization technique. In this paper, different normalization methods are used on time series data before feeding the data into the DNN model and we try to find out the impact of each normalization technique on DNN to forecast the time series. Here the Deep Recurrent Neural Network (DRNN) is used to predict the closing index of Bombay Stock Exchange (BSE) and New York Stock Exchange (NYSE) by using BSE and NYSE time series data.

Staircase Attention for Recurrent Processing of Sequences

Attention mechanisms have become a standard tool for sequence modeling tasks, in particular by stacking self-attention layers over the entire input sequence as in the Transformer architecture. In this work we introduce a novel attention procedure called staircase attention that, unlike self-attention, operates across the sequence (in time) recurrently processing the input by adding another step of processing. A step in the staircase comprises of backward tokens (encoding the sequence so far seen) and forward tokens (ingesting a new part of the sequence), or an extreme Ladder version with a forward step of zero that simply repeats the Transformer on each step of the ladder, sharing the weights. We thus describe a family of such models that can trade off performance and compute, by either increasing the amount of recurrence through time, the amount of sequential processing via recurrence in depth, or both. Staircase attention is shown to be able to solve tasks that involve tracking that conventional Transformers cannot, due to this recurrence. Further, it is shown to provide improved modeling power for the same size model (number of parameters) compared to self-attentive Transformers on large language modeling and dialogue tasks, yielding significant perplexity gains.

Accelerating Neural Network Training with Distributed Asynchronous and Selective Optimization (DASO)

With increasing data and model complexities, the time required to train neural networks has become prohibitively large. To address the exponential rise in training time, users are turning to data parallel neural networks (DPNN) to utilize large-scale distributed resources on computer clusters. Current DPNN approaches implement the network parameter updates by synchronizing and averaging gradients across all processes with blocking communication operations. This synchronization is the central algorithmic bottleneck. To combat this, we introduce the Distributed Asynchronous and Selective Optimization (DASO) method which leverages multi-GPU compute node architectures to accelerate network training. DASO uses a hierarchical and asynchronous communication scheme comprised of node-local and global networks while adjusting the global synchronization rate during the learning process. We show that DASO yields a reduction in training time of up to 34% on classical and state-of-the-art networks, as compared to other existing data parallel training methods.

UltraBot: Autonomous Mobile Robot for Indoor UV-C Disinfection with Non-trivial Shape of Disinfection Zone

The paper focuses on the development of an autonomous disinfection robot UltraBot to reduce COVID-19 transmission along with other harmful bacteria and viruses. The motivation behind the research is to develop such a robot that is capable of performing disinfection tasks without the use of harmful sprays and chemicals that can leave residues and require airing the room afterward for a long time. UltraBot technology has the potential to offer the most optimal autonomous disinfection performance along with taking care of people, keeping them from getting under the UV-C radiation. The paper highlights UltraBot's mechanical and electrical design as well as disinfection performance. The conducted experiments demonstrate the effectiveness of robot disinfection ability and actual disinfection area per each side with UV-C lamp array. The disinfection effectiveness results show actual performance for the multi-pass technique that provides 1-log reduction with combined direct UV-C exposure and ozone-based air purification after two robot passes at a speed of 0.14 m/s. This technique has the same performance as ten minutes static disinfection. Finally, we have calculated the non-trivial form of the robot disinfection zone by two consecutive experiment to produce optimal path planning and to provide full disinfection in selected areas.