Neural architecture search (NAS) has become a key component of AutoML and a standard tool to automate the design of deep neural networks. Recently, training-free NAS as an emerging paradigm has successfully reduced the search costs of standard training-based NAS by estimating the true architecture performance with only training-free metrics. Nevertheless, the estimation ability of these metrics typically varies across different tasks, making it challenging to achieve robust and consistently good search performance on diverse tasks with only a single training-free metric. Meanwhile, the estimation gap between training-free metrics and the true architecture performances limits training-free NAS to achieve superior performance. To address these challenges, we propose the robustifying and boosting training-free NAS (RoBoT) algorithm which (a) employs the optimized combination of existing training-free metrics explored from Bayesian optimization to develop a robust and consistently better-performing metric on diverse tasks, and (b) applies greedy search, i.e., the exploitation, on the newly developed metric to bridge the aforementioned gap and consequently to boost the search performance of standard training-free NAS further. Remarkably, the expected performance of our RoBoT can be theoretically guaranteed, which improves over the existing training-free NAS under mild conditions with additional interesting insights. Our extensive experiments on various NAS benchmark tasks yield substantial empirical evidence to support our theoretical results.

Understanding how the predictions of deep learning models are formed during the training process is crucial to improve model performance and fix model defects, especially when we need to investigate nontrivial training strategies such as active learning, and track the root cause of unexpected training results such as performance degeneration. In this work, we propose a time-travelling visual solution DeepVisualInsight (DVI), aiming to manifest the spatio-temporal causality while training a deep learning image classifier. The spatio-temporal causality demonstrates how the gradient-descent algorithm and various training data sampling techniques can influence and reshape the layout of learnt input representation and the classification boundaries in consecutive epochs. Such causality allows us to observe and analyze the whole learning process in the visible low dimensional space. Technically, we propose four spatial and temporal properties and design our visualization solution to satisfy them. These properties preserve the most important information when projecting and inverse-projecting input samples between the visible low-dimensional and the invisible high-dimensional space, for causal analyses. Our extensive experiments show that, comparing to baseline approaches, we achieve the best visualization performance regarding the spatial/temporal properties and visualization efficiency. Moreover, our case study shows that our visual solution can well reflect the characteristics of various training scenarios, showing good potential of DVI as a debugging tool for analyzing deep learning training processes.

Medication recommendation is a crucial task for intelligent healthcare systems. Previous studies mainly recommend medications with electronic health records (EHRs). However, some details of interactions between doctors and patients may be ignored or omitted in EHRs, which are essential for automatic medication recommendation. Therefore, we make the first attempt to recommend medications with the conversations between doctors and patients. In this work, we construct DIALMED, the first high-quality dataset for medical dialogue-based medication recommendation task. It contains 11, 996 medical dialogues related to 16 common diseases from 3 departments and 70 corresponding common medications. Furthermore, we propose a Dialogue structure and Disease knowledge aware Network (DDN), where a QA Dialogue Graph mechanism is designed to model the dialogue structure and the knowledge graph is used to introduce external disease knowledge. The extensive experimental results demonstrate that the proposed method is a promising solution to recommend medications with medical dialogues. The dataset and code are available at https://github.com/f-window/DialMed.

A modified two-group resource allocation scheme is tested with three channel ordering schemes, a hard decision parallel interference cancellation (HDPIC) scheme and a successive interference cancellation (SIC) scheme to improve the data rate. Channel ordering ensures channels with the lowest energy requirement are loaded with the higher data rate in the two-group resource allocation scheme but may yield an increased computational load, which affects the feasibility of hardware implementation such as in Femtocells. Furthermore, channel ordering is also needed in SIC schemes to reduce error propagation loss, hence improving detection reliability.

This paper studies an efficient radio resource allocation method for the multicode transmission of High-Speed Downlink Packet Access system. The current HSDPA system allocates equal data rates to multiple channels, causing an inefficient use of radio resources in multipath fading channel. To maximize the system throughput subject to limited radio resources, this paper proposes a computing-efficient 2 group allocation method, which assigns two distinct data rates to 2 groups of channels respectively. The two data rates and the division of the channels can be derived iteratively. The 2 group allocation method significantly outperforms the link adaptation method in the HSDPA system in system throughput, especially when the system operates in high SNR region.

The current high speed downlink packet access (HSDPA) system with the quadrature amplitude modulation (QAM) allocates equal data rates to multiple codes for link adaptation which causes large quantization loss in the total data rates. The 2 group allocation method proposed in significantly reduces the quantization loss by optimally adjusting the data rates over each channel for the QAM-modulated underloaded system. This paper applies this rate adaptive throughput maximization problem to the pulse amplitude modulated (PAM) overloaded systems. The proposed system uses complex Grass-mannian codes as the spreading sequences and a new complex conjugate concatenation technique for the receiver to further reduce the quantization loss. The numerical results show that the proposed PAM-modulated system employing the 2 group allocation scheme increases the total data rate approximately by 30 - 100% compared with the link adaptation method currently used in the HSDPA system.