Explicit Duration Research Articles

Lokatt: a hybrid DNA nanopore basecaller with an explicit duration hidden Markov model and a residual LSTM network

BackgroundBasecalling long DNA sequences is a crucial step in nanopore-based DNA sequencing protocols. In recent years, the CTC-RNN model has become the leading basecalling model, supplanting preceding hidden Markov models (HMMs) that relied on pre-segmenting ion current measurements. However, the CTC-RNN model operates independently of prior biological and physical insights.ResultsWe present a novel basecaller named Lokatt: explicit duration Markov model and residual-LSTM network. It leverages an explicit duration HMM (EDHMM) designed to model the nanopore sequencing processes. Trained on a newly generated library with methylation-free Ecoli samples and MinION R9.4.1 chemistry, the Lokatt basecaller achieves basecalling performances with a median single read identity score of 0.930, a genome coverage ratio of 99.750%, on par with existing state-of-the-art structure when trained on the same datasets.ConclusionOur research underlines the potential of incorporating prior knowledge into the basecalling processes, particularly through integrating HMMs and recurrent neural networks. The Lokatt basecaller showcases the efficacy of a hybrid approach, emphasizing its capacity to achieve high-quality basecalling performance while accommodating the nuances of nanopore sequencing. These outcomes pave the way for advanced basecalling methodologies, with potential implications for enhancing the accuracy and efficiency of nanopore-based DNA sequencing protocols.

Application-Layer DDoS Attack Detection Using Explicit Duration Recurrent Network-Based Application-Layer Protocol Communication Models

Existing application-layer distributed denial of service (AL-DDoS) attack detection methods are mainly targeted at specific attacks and cannot effectively detect other types of AL-DDoS attacks. This study presents an application-layer protocol communication model for AL-DDoS attack detection, based on the explicit duration recurrent network (EDRN). The proposed method includes model training and AL-DDoS attack detection. In the AL-DDoS attack detection phase, the output of each observation sequence is updated in real time. The observation sequences are based on application-layer protocol keywords and time intervals between adjacent protocol keywords. Protocol keywords are extracted based on their identification using regular expressions. Experiments are conducted using datasets collected from a real campus network and the CICDDoS2019 dataset. The results of the experiments show that EDRN is superior to several popular recurrent neural networks in accuracy, F1, recall, and loss values. The proposed model achieves an accuracy of 0.996, F1 of 0.992, recall of 0.993, and loss of 0.041 in detecting HTTP DDoS attacks on the CICDDoS2019 dataset. The results further show that our model can effectively detect multiple types of AL-DDoS attacks. In a comparison test, the proposed method outperforms several state-of-the-art approaches.

EDRN-based propagation model for popular microblog information detection

Microblogging platforms have become popular for sharing knowledge and communicating with others. Rumours and current trends can be identified promptly if popular information trends on these platforms can be detected in the early stages of dissemination. Microblog information propagation is primarily based on forwarders; the forwarded information is more likely to become popular if the forwarder's influence is greater and the time interval between two adjacent forwarders is shorter. Based on the explicit duration recurrent network (EDRN), a recurrent neural network, a new microblog information-propagation model for the early detection of popular information was presented. In this model, the observation is constructed based on the forwarder's influence level and the time interval between two adjacent forwarders. The influence level of the forwarder was estimated using the support vector machine classifier. To evaluate this model, experiments were conducted using Twitter and Weibo datasets. In the experiments, EDRN outperformed several popular recurrent neural networks, hidden semi-Markov models, and hidden Markov models, particularly when the training set was small. The timeliness evaluation results demonstrated that the model can effectively detect popular information at an early stage.

Real-time monitoring of reaction mechanisms from spectroscopic data using hidden semi-Markov models for mode identification

In this work, we present a framework for process monitoring focusing on the dynamics of reaction mechanisms based purely on online spectroscopic data. This is accomplished by developing an explicit duration hidden semi-Markov model (HSMM) that is used to monitor changes in reaction mechanisms with changing temperatures in a complex reacting system by dynamically identifying groups of spectroscopic samples that belong to a mode, and the mode duration of the reaction mechanism associated with the samples. An expectation maximization algorithm is used for parameter re-estimation, and Viterbi state decoding is used to identify the most likely sequence of hidden states/modes that may have generated the observed sequence of spectra. The reaction mechanism associated with samples of a mode is then deduced by extracting latent features among spectra of the mode and learning a probabilistic graphical structure among the features using Bayesian networks, which represent a network or mechanism of hypothesized reactions. The technique is demonstrated on case studies related to the partial upgrading of bitumen using thermochemical conversion based on the acquisition of Fourier transform infrared spectroscopic data. This system is complex enough that prior information regarding both species and reactions is unavailable. Both offline and online monitoring are implemented for mode identification, and the technique provides monitoring of the multi-modal process and, at the same time, provides insight into the chemistry specific to each mode, which makes it useful both for process control and fundamental studies into process chemistry. Synthetic dynamic spectral data, that is derived through interpolation from real spectral data obtained at various static conditions of temperature and residence time, is used in the study.

Explicit Duration Recurrent Networks.

Recurrent neural networks (RNNs) can be used to operate over sequences of vectors and have been successfully applied to a variety of problems. However, it is hard to use RNNs to model the variable dwell time of the hidden state underlying an input sequence. In this article, we interpret the typical RNNs, including original RNN, standard long short-term memory (LSTM), peephole LSTM, projected LSTM, and gated recurrent unit (GRU), using a slightly extended hidden Markov model (HMM). Based on this interpretation, we are motivated to propose a novel RNN, called explicit duration recurrent network (EDRN), analog to a hidden semi-Markov model (HSMM). It has a better performance than conventional LSTMs and can explicitly model any duration distribution function of the hidden state. The model parameters become interpretable and can be used to infer many other quantities that the conventional RNNs cannot obtain. Therefore, EDRN is expected to extend and enrich the applications of RNNs. The interpretation also suggests that the conventional RNNs, including LSTM and GRU, can be made small modifications to improve their performance without increasing the parameters of the networks.

Implicit Versus Explicit Timing-Separate or Shared Mechanisms?

Time implicitly shapes cognition, but time is also explicitly represented, for instance, in the form of durations. Parsimoniously, the brain could use the same mechanisms for implicit and explicit timing. Yet, the evidence has been equivocal, revealing both joint versus separate signatures of timing. Here, we directly compared implicit and explicit timing using magnetoencephalography, whose temporal resolution allows investigating the different stages of the timing processes. Implicit temporal predictability was induced in an auditory paradigm by a manipulation of the foreperiod. Participants received two consecutive task instructions: discriminate pitch (indirect measure of implicit timing) or duration (direct measure of explicit timing). The results show that the human brain efficiently extracts implicit temporal statistics of sensory environments, to enhance the behavioral and neural responses to auditory stimuli, but that those temporal predictions did not improve explicit timing. In both tasks, attentional orienting in time during predictive foreperiods was indexed by an increase in alpha power over visual and parietal areas. Furthermore, pretarget induced beta power in sensorimotor and parietal areas increased during implicit compared to explicit timing, in line with the suggested role for beta oscillations in temporal prediction. Interestingly, no distinct neural dynamics emerged when participants explicitly paid attention to time, compared to implicit timing. Our work thus indicates that implicit timing shapes the behavioral and sensory response in an automatic way and is reflected in oscillatory neural dynamics, whereas the translation of implicit temporal statistics to explicit durations remains somewhat inconclusive, possibly because of the more abstract nature of this task.

Evaluation of Tacotron Based Synthesizers for Spanish and Basque

In this paper, we describe the implementation and evaluation of Text to Speech synthesizers based on neural networks for Spanish and Basque. Several voices were built, all of them using a limited number of data. The system applies Tacotron 2 to compute mel-spectrograms from the input sequence, followed by WaveGlow as neural vocoder to obtain the audio signals from the spectrograms. The limited number of data used for training the models leads to synthesis errors in some sentences. To automatically detect those errors, we developed a new method that is able to find the sentences that have lost the alignment during the inference process. To mitigate the problem, we implemented a guided attention providing the system with the explicit duration of the phonemes. The resulting system was evaluated to assess its robustness, quality and naturalness both with objective and subjective measures. The results reveal the capacity of the system to produce good quality and natural audios.

Tonal cues to prosodic structure in rate-dependent speech perception.

This study explores how listeners integrate tonal cues to prosodic structure with their perception of local speech rate and consequent interpretation of durational cues. In three experiments, we manipulate the pitch and duration of speech segments immediately preceding a target sound along a vowel duration continuum (cueing coda stop voicing), testing how listeners' categorization of vowel duration shifts based on temporal and tonal context. We find that listeners perceive the presence of a phrasal boundary tone on a lengthened syllable as signaling a slowdown in speech rate, shifting perception of vowel duration, with effects that are additive when crossed in a 2 × 2 (pitch × duration) design. However, an asymmetrical effect of pitch and duration is found in an explicit duration judgement task in which listeners judge how long a pre-target syllable sounds to them. In explicit rate judgement, only durational information is consequential, unlike the categorization task, suggesting that integration of tonal and durational prosodic cues in rate-dependent perception is limited to implicit processing of speech rate. Results are discussed in terms of linguistic information in rate-dependent speech processing, the integration of prosodic cues, and implicit and explicit rate processing tasks.

Quantitative Analysis of Forest Fires in Southeastern Australia Using SAR Data

Prescribed burning is a common strategy for minimizing forest fire risk. Fire is introduced under specific environmental conditions, with explicit duration, intensity, and rate of spread. Such conditions deviate from those encountered during the fire season. Prescribed burns mostly affect surface fuels and understory vegetation, an outcome markedly different when compared to wildfires. Data on prescribed burning are crucial for evaluating whether land management targets have been reached. This research developed a methodology to quantify the effects of prescribed burns using multi-temporal Sentinel-1 Synthetic Aperture Radar (SAR) imagery in the forests of southeastern Australia. C-band SAR datasets were specifically used to statistically explore changes in radar backscatter coefficients with the intensity of prescribed burns. Two modeling approaches based on pre- and post-fire ratios were applied for evaluating prescribed burn impacts. The effects of prescribed burns were documented with an overall accuracy of 82.3% using cross-polarized backscatter (VH) SAR data under dry conditions. The VV polarization indicated some potential to detect burned areas under wet conditions. The findings in this study indicate that the C-band SAR backscatter coefficient has the potential to evaluate the effectiveness of prescribed burns due to its sensitivity to changes in vegetation structure.

Online Unsupervised Occupancy Anticipation System Applied to Residential Heat Load Management

Human preferences and lifestyles significantly impact buildings' energy consumption. Consequently, a better understanding of occupants' behavior is crucial to decrease energy consumption and maintain occupants' comfort. Occupant-centric control (OCC) strategies are effective approaches to fulfil such a purpose. As such, occupancy detection and prediction are of prime importance, particularly to manage Electric Space Heating (ESH) systems, due to the relatively slow dynamics of the temperature in dwellings. This paper proposes an Explicit Duration Hidden Markov Model (EDHMM) for unsupervised online presence detection and a hazard-based approach for occupancy prediction. Moreover, a control strategy using a cost function, weighted by occupancy predictions, and a load-shifting strategy based on time-varying electricity price are put forward. This work initially validates the consistency of the proposed approach by using synthetic data generated by a Monte Carlo simulation. Subsequently, the performance of our framework is compared with previous methods presented in the literature through experimental validation. Results demonstrate that the proposed EDHMM approach is efficient in detecting occupancy states. Besides, the results of the field implementation show the potential of the proposed control strategy to preserve occupants' thermal comfort while decreasing the heating energy consumption.

The internal representation of temporal orienting: A temporal pulse-accumulation and attentional-gating-based account.

Timing can be processed explicitly or implicitly. Temporal orienting is a typical implicit timing through which we can anticipate and prepare an optimized response to forthcoming events. It is, however, not yet clear whether mechanisms such as temporal-pulse accumulation and attentional gating (more attention, more accumulated temporal pulses) underly the internal representations of temporal orienting, as in explicit timing. To clarify this, a dual-task paradigm, consisting of a temporal orienting and an interference task, was adopted. Consistent with the temporal-pulse-accumulation and attentional-gating model, reaction times to the target detection of temporal orienting increased as the interference stimuli were temporally closer to the target, i.e., a location effect for temporal orienting. This effect is likely due to attention being diverted away from temporal orienting to monitor the occurrence of the interference stimuli for a longer time, resulting in greater temporal pulse loss and less accurate temporal orienting for conditions with later interference stimuli. The temporal-pulse-accumulation aspect in temporal orienting received further support by taking an explicit duration reproduction (containing a second temporal-pulse accumulation) as the interference task. On the one hand, temporal orienting became less accurate with increased temporal-pulse-accumulation overlaps between the dual tasks; on the other hand, two-way (one for temporal orienting and the other for duration reproduction), rather than one-way, location effects were observed, implying processing conflicts between the two temporal-pulse accumulations. Taken together, these results suggest that implicit and explicit timing may share common mechanisms upon internal temporal representations.

Machine Learning Based Network Analysis Using Millimeter-Wave Narrow-Band Energy Traces

Next-generation wireless networks promise to provide extremely high data rates, especially exploiting the so-called millimeter-wave frequency range. Gaining information from spectrum usage is becoming important to provide smart adaptation capabilities to future network protocol stacks. Issues such as deafness, misaligned antennas, or blockage may severely impact network performance, and their identification is crucial. Despite the complexity of full analytical models, machine learning techniques are progressively being considered to improve spectrum usage at higher layers. In this paper, we design a signal processing technique that uses narrowband physical layer energy traces, obtained from one or multiple channel sniffers. The proposed technique utilizes a combination of template matching and an Explicit Duration Hidden Markov Model (EDHMM) to correctly classify frames, while coping with the non-stationarity of the traces. This leads to a protocol level monitor that does not need to decode the channel at the physical layer, but just infers the type of packets that are exchanged based on sub-sampled energy traces. The performance of this framework is evaluated using off-the-shelf mm-wave wireless devices, quantifying its detection performance in the presence of one or multiple sniffers, and assessing the impact of physical layer parameters such as noise power and signal levels.

Estimating prescribed fire impacts and post-fire tree survival in eucalyptus forests of Western Australia with L-band SAR data

Prescribed burning is a technique commonly applied to reduce fire risk. Fire is introduced under specific environmental conditions, with explicit duration, intensity and rate of spread. Such conditions deviate from those encountered during the fire season. As a result, prescribed burns mostly affect surface fuels and understorey vegetation, an outcome markedly different when compared to wildfires.Information on prescribed burning outcomes is important for evaluating the extent to which land management objectives have been achieved. This study focused on the ability of Radar Burn Ratio (RBR), a Synthetic Aperture Radar (SAR)-based multi-temporal index, to appraise the outcome of prescribed burns in eucalypt forests of Western Australia. L-band datasets acquired by the Phased Array type L-band Synthetic Aperture Radar (PALSAR-2) sensor were used to relate changes in radar backscatter coefficient with the effects of prescribed burning. Two modelling approaches based on post- to pre-fire ratios were used to evaluate prescribed burn impacts. Post-fire tree survival in areas affected by high fire impacts was inferred by assessing the recovery rates of the backscatter coefficient for the first year after the burn.The impact of prescribed burns was estimated with relatively high overall accuracy (79%) using cross-polarized backscatter (HV) under dry conditions. The HH polarization and the Normalized Difference Backscatter Intensity (NDBI) showed certain ability to detect burned areas under wet conditions. The results suggest that L-band backscatter coefficient may contribute to evaluating the success of prescribed burns in Western Australian eucalypt forests through its sensitivity to changes in vegetation structure. In addition, tree survival (as opposed to mortality) and recovery towards the pre-fire state might be inferred by analysing short-term post-fire trends of the backscatter coefficient in areas affected by high fire impacts.

The challenges of assessing effectiveness of lacosamide using electronic medical record databases

PurposeElectronic health record (EHR) databases are a potential source for conducting research to generate real world evidence on patient outcomes. The objective of the study was to evaluate the feasibility of using EHR data to assess seizure outcomes in patients treated with lacosamide (LCM) monotherapy. MethodsThis was a retrospective cohort study conducted using the Optum clinical EHR database. The study sample comprised patients ≥17 years of age with epilepsy or seizures and treated with LCM monotherapy between 1 January 2009 and 31 December 2013. Structured and unstructured data from prescribed medication and abstracted physician note records were used to identify patients with epilepsy treated with LCM monotherapy and measure seizure frequency outcomes. The index date was the first date of LCM monotherapy, with a 6-month baseline period. Patients were observed for up to 12 months beginning on the index date (follow-up period). The EHR data were not sufficient to compute days supply and explicit duration of LCM and other antiepileptic drug (AED) therapies; therefore, LCM monotherapy was estimated from prescription dates of AEDs. Outcomes were change in seizures per month or change in seizure frequency category from baseline to follow-up. Descriptive statistics were used to describe baseline characteristics and study outcomes. ResultsA total of 10,988 patients with at least one LCM prescription were identified during the study period, 470 of whom met all the selection criteria and were included in the study sample. Although many patients had abstracted physician note records that referred to their seizures, only 3.2% of the patients had seizure frequency information that could be used to quantify the number of seizures per month in both the baseline and follow-up periods; thus, this information could not be used to assess the effectiveness of LCM monotherapy on seizure outcomes. ConclusionLacosamide monotherapy effectiveness was not estimated because the EHR prescription record data did not have sufficient information on days supply. Additionally, most patients' records did not contain adequate information to allow for evaluation of quantitative changes in seizure frequency based on the number of seizures per month. More studies are needed to validate these study findings.

Testing Applicability of Virtual Stochastic Sensors for Non-Intrusive Appliance Load Monitoring

Non-intrusive appliance load monitoring (NIALM) aims at reconstructing the electricity consumption of household appliances based only on the cumulative consumption data collected via a smart meter. Various approaches have been proposed to perform NIALM including various types of Hidden Markov Model (HMM) offspring. Most do not consider the explicit duration of an appliance activation. Virtual stochastic sensors (VSS) and their underlying Hidden non-Markovian Models (HnMM) can include explicit process durations. This paper tests whether VSS can solve the NIALM task, and analyzes the methods reconstruction accuracy on the publicly available SMART* data set. Models of the household appliances with different inherent states are automatically extracted from the appliance data. The combined model, including a subset of the appliances, is then used to disaggregate the cumulative energy consumption data. Experiments show a reconstruction accuracy of up to 90% with appropriate method parameters, showing that VSS can compete with existing NIALM approaches.

Action recognition by Latent Duration Model

Temporal structure has attracted lots of research interests due to its ability to model complex pattern for effective action recognition. Most existing methods exploit temporal structure only in a fixed scale or implicit multiple scales. Although recently some methods attempt towards exploiting the temporal structure and relationship by using the durations of action primitives, they cannot effectively conduct the action recognition and discriminative segments discovery simultaneously. In this paper, we propose a novel action recognition method, named Latent Duration Model (LDM), which is a temporal variant of Deformable Part Model (DPM) with explicit durations and temporal ordering constraints. Three types of latent variables are introduced into LDM. Latent duration variables are used to accommodate intra-class temporal scale variation. Latent location variables and latent representation variables are utilized to help search the most discriminative segments in the durations. For temporal structure and relationship, our model takes into account both temporal order and duration changes between consecutive parts, which are robust and flexible to the variety in motion speeds and view angel changes of action videos. Thus, not only discriminative parts with adaptive durations but also robust pairwise relationship is automatically discovered by our model. The experimental results on Olympic Sports, Hollywood2, UCF50 and HMDB51 datasets show the effectiveness of our proposed model.

Unsupervised and phonologically controlled interpolation of Austrian German language varieties for speech synthesis

This paper presents an unsupervised method that allows for gradual interpolation between language varieties in statistical parametric speech synthesis using Hidden Semi-Markov Models (HSMMs). We apply dynamic time warping using Kullback–Leibler divergence on two sequences of HSMM states to find adequate interpolation partners. The method operates on state sequences with explicit durations and also on expanded state sequences where each state corresponds to one feature frame. In an intelligibility and dialect rating subjective evaluation of synthesized test sentences, we show that our method can generate intermediate varieties for three Austrian dialects (Viennese, Innervillgraten, Bad Goisern). We also provide an extensive phonetic analysis of the interpolated samples. The analysis includes input-switch rules, which cover historically different phonological developments of the dialects versus the standard language; and phonological processes, which are phonetically motivated, gradual, and common to all varieties. We present an extended method which linearly interpolates phonological processes but uses a step function for input-switch rules. Our evaluation shows that the integration of this kind of phonological knowledge improves dialect authenticity judgment of the synthesized speech, as performed by dialect speakers. Since gradual transitions between varieties are an existing phenomenon, we can use our methods to adapt speech output systems accordingly.

A Logical Hierarchical Hidden Semi-Markov Model for Team Intention Recognition

Intention recognition is significant in many applications. In this paper, we focus on team intention recognition, which identifies the intention of each team member and the team working mode. To model the team intention as well as the world state and observation, we propose a Logical Hierarchical Hidden Semi-Markov Model (LHHSMM), which has advantages of conducting statistical relational learning and can present a complex mission hierarchically. Additionally, the LHHSMM explicitly models the duration of team working mode, the intention termination, and relations between the world state and observation. A Logical Particle Filter (LPF) algorithm is also designed to infer team intentions modeled by the LHHSMM. In experiments, we simulate agents’ movements in a combat field and employ agents’ traces to evaluate performances of the LHHSMM and LPF. The results indicate that the team working mode and the target of each agent can be effectively recognized by our methods. When intentions are interrupted within a high probability, the LHHSMM outperforms a modified logical hierarchical hidden Markov model in terms of precision, recall, andF-measure. By comparing performances of LHHSMMs with different duration distributions, we prove that the explicit duration modeling of the working mode is effective in team intention recognition.

A Comparison of DSM-IV-TR and DSM-5 Definitions for Sexual Dysfunctions: Critiques and Challenges

The diagnostic criteria of sexual dysfunctions (SDs) are paramount for the development of sexual medicine as reliable diagnoses are essential to guide treatment plans. Prior Diagnostic and Statistical Manual of Mental Disorders (DSM) classifications based definitions of SD mostly on expert opinions and included imprecise terms. The validity of diagnoses of SD has only recently been challenged, and efforts are made to make more operational definitions. This paper aims to compare and contrast the recently released Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (DSM-5) diagnostic criteria of SD with that of Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition Text Revision (DSM-IV-TR) and explains the rationale for making changes in the new DSM-5. It also aims to address some issues to be considered further for the future. Online proposed American Psychiatric Association website DSM-5, the new released DSM-5, and DSM-IV-TR diagnostic criteria for SD were thoroughly inspected, and an extensive literature search was performed for comparative reasons. Changes in diagnostic criteria of DSM-5 were detected, and DSM-IV-TR and DSM-5 diagnostic criteria for SD were compared and contrasted. Diagnostic criteria were more operationalized, and explicit duration and frequency criteria were set up in DSM-5 for purposes of good clinical research. Classifications based on simple linear sexual response were abandoned, and diagnostic classifications were separately made for males and females. Desire and arousal disorders in women were merged. Drifting apart from linear sexual response cycle may be an advancement in establishing specific diagnostic criteria for different genders. However, it is still a question of debate whether there is enough evidence to lump sexual interest and arousal disorders in females. Making more precise definitions is important to differentiate disorders from other transient conditions. However, there is still room to improve our definitions and find a way to include gay and lesbian individuals. Further discussions and debates are expected to be continued in the future.

Time Petri Nets with Action Duration

The design of real-time systems needs a high-level specification model supporting at the same time timing constraints and actions duration. The authors introduce in this paper an extension of Petri Nets called Time Petri Nets with Action Duration (DTPN) where time is associated with transitions. In DTPN, the firing of transitions is bound to a time interval and transitions represent actions which have explicit durations. The authors give an operational semantics for DTPN in terms of Durational Action Timed Automata (DATA). DTPN considers both timing constraints and durations under a true-concurrency semantics with an aim of better expressing concurrent and parallel behaviours of real-time systems.