Generating explanations for neural networks has become crucial for their applications in real-world with respect to reliability and trustworthiness. In natural language processing, existing methods usually provide important features which are words or phrases selected from an input text as an explanation, but ignore the interactions between them. It poses challenges for humans to interpret an explanation and connect it to model prediction. In this work, we build hierarchical explanations by detecting feature interactions. Such explanations visualize how words and phrases are combined at different levels of the hierarchy, which can help users understand the decision-making of blackbox models. The proposed method is evaluated with three neural text classifiers (LSTM, CNN, and BERT) on two benchmark datasets, via both automatic and human evaluations. Experiments show the effectiveness of the proposed method in providing explanations that are both faithful to models and interpretable to humans.
Motivation Genomic region sets summarize functional genomics data and define locations of interest in the genome such as regulatory regions or transcription factor binding sites. The number of publicly available region sets has increased dramatically, leading to challenges in data analysis. Results We propose a new method to represent genomic region sets as vectors, or embeddings, using an adapted word2vec approach. We compared our approach to two simpler methods based on interval unions or term frequency-inverse document frequency and evaluated the methods in three ways: First, by classifying the cell line, antibody, or tissue type of the region set; second, by assessing whether similarity among embeddings can reflect simulated random perturbations of genomic regions; and third, by testing robustness of the proposed representations to different signal thresholds for calling peaks. Our word2vec-based region set embeddings reduce dimensionality from more than a hundred thousand to 100 without significant loss in classification performance. The vector representation could identify cell line, antibody, and tissue type with over 90% accuracy. We also found that the vectors could quantitatively summarize simulated random perturbations to region sets and are more robust to subsampling the data derived from different peak calling thresholds. Our evaluations demonstrate that the vectors retain useful biological information in relatively lower-dimensional spaces. We propose that vector representation of region sets is a promising approach for efficient analysis of genomic region data. Availability https://github.com/databio/regionset-embedding
A visible light communication (VLC) system can adopt multi-color light emitting diode (LED) arrays to support multiple users. In this paper, a multi-layer coding and constrained partial group decoding (CPGD) method is proposed to tackle strong color interference and increase the system throughput. After channel model formulation, user information rates are allocated and decoding order for all the received data layers is obtained by solving a max-min fairness problem using a greedy algorithm. An achievable rate is derived under the truncated Gaussian input distribution. To reduce the decoding complexity, a map on the decoding order and rate allocation is constructed for all positions of interest on the receiver plane and its size is reduced by a classification-based algorithm. Meanwhile, the symmetrical geometry of LED arrays is exploited. Finally, the transmitter-user association problem is formulated and solved by a genetic algorithm. It is observed that the system throughput increases as the receivers are slightly misaligned with corresponding LED arrays due to the reduced interference level, but decreases afterwards due to the weakened link gain. Index TermsVisible light communication, multi-layer transmission, constrained partial group decoder, interference cancellation, transmitter-user association. 2 I. INTRODUCTION With the increasing amount of data transmitted via wireless links, the spectrum shortage has become a critical problem for the next generation communication systems. Visible light communication (VLC) has emerged as a competent supplement [1], [2] to radio-frequency communication due to its unique advantages, such as large bandwidth and free of electromagnetic radiation. Combining with its other advantages, such as simple transceiver structures due to its intensity modulation-direct detection (IM-DD) method, VLC has attracted extensive research attentions in recent years [3]-[6].Light emitting diode (LED) array can be adopted to provide required illumination and simultaneously increase the system's data rate [7], [8]. In a multi-user scenario, where each LED and photodiode (PD) pair serves one user, the PD receives not only the desired signal but also the signals from adjacent LEDs as interferences. It is typical that these interferences are strong and thus treating them as noise will penalize the system throughput or even cause a decoding failure. To tackle this problem, interference can be suppressed by designing a resource allocation scheme [9] or aligned [10], [11] via transmitter-side signal processing.One the other hand, it is generally accepted that while a receiver is not interested in decoding the messages from interferers, decoding them is often beneficial for recovering the desired message [12]. Multi-layer coding and group decoder [13] provide an interference cancellation framework that exploits this benefit. At the transmitter, each data stream is split into multiple layers that are encoded individually using independent codebooks, followed by an LED sending the superposition of t...
Generating explanations for neural networks has become crucial for their applications in real-world with respect to reliability and trustworthiness. In natural language processing, existing methods usually provide important features which are words or phrases selected from an input text as an explanation, but ignore the interactions between them. It poses challenges for humans to interpret an explanation and connect it to model prediction. In this work, we build hierarchical explanations by detecting feature interactions. Such explanations visualize how words and phrases are combined at different levels of the hierarchy, which can help users understand the decision-making of blackbox models. The proposed method is evaluated with three neural text classifiers (LSTM, CNN, and BERT) on two benchmark datasets, via both automatic and human evaluations. Experiments show the effectiveness of the proposed method in providing explanations that are both faithful to models and interpretable to humans.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.