Formal assessment of reliability specifications in embedded cyber-physical systems

Reliability has become an integral component of the design intent of embedded cyber-physical systems. Safety-critical embedded systems are designed with specific reliability targets, and design practices include the appropriate allocation of both spatial and temporal redundancies in the implementation to meet such requirements. With increasing complexity of such systems and considering the large number of components in such systems, redundancy allocation requires a formal scientific basis. In this work, we profess the analysis of the redundancy requirement upfront with the objective of making it an integral part of the specification. The underlying problem is one of synthesizing a formal specification with built-in redundancy artifacts, from the formal properties of the error-free system, the error probabilities of the control components, and the reliability target. We believe that upfront formal analysis of redundancy requirements is important in budgeting the resource requirements from a cost versus reliability perspective. Several case-studies from the automotive domain highlight the efficacy of our proposal.     

The Responses of Health Systems to Marital Sexual Violence – A Perspective from Southern India

ABSTRACT

Marital sexual violence is a serious problem in India. However, marital rape and most other forms of marital sexual violence are not criminalized in the country. This qualitative study with healthcare providers (physicians and nurses), lawyers, members of a non-profit organization that offers domestic violence support services out of a hospital, a journalist and two lawyers suggest that the majority of care providers recognize that marital sexual violence is a serious concern, and agree that health systems can play a vital role in addressing both the immediate biomedical concerns of survivors, and also overall well-being. The study reveals several systemic and internal factors that hinder responsiveness of Indian health institutions. These include, the absence of a protocol or uniform internal guidelines in most hospitals, a lack of screening programs to capture the presence of violence occurring when patients visit hospitals for other reasons, a large patient load, a lack of gender sensitivity in providers’ education, an absence of domestic violence shelters, and poor interlinkages between support services such as counseling and legal services. Notwithstanding these challenges, some exemplar physicians and nurses assist survivors of sexual violence, sometimes risking their own safety. The Indian health system can be made more responsive, provided these factors are addressed, and crucially budgets are allocated for interventions.     

Memory systems and the control of skilled action

ABSTRACT

In keeping with the dominant view that skills are largely automatic, the standard view of memory systems distinguishes between a representational declarative system associated with cognitive processes and a performance-based procedural system. The procedural system is thought to be largely responsible for the performance of well-learned skilled actions. Here we argue that most skills do not fully automate, which entails that the declarative system should make a substantial contribution to skilled performance. To support this view, we review evidence showing that the declarative system does indeed play a number of roles in skilled action.     

Speaker identification using multi-modal i-vector approach for varying length speech in voice interactive systems

The development in the interface of smart devices has lead to voice interactive systems. An additional step in this direction is to enable the devices to recognize the speaker. But this is a challenging task because the interaction involves short duration speech utterances. The traditional Gaussian mixture models (GMM) based systems have achieved satisfactory results for speaker recognition only when the speech lengths are sufficiently long. The current state-of-the-art method utilizes i-vector based approach using a GMM based universal background model (GMM-UBM). It prepares an i-vector speaker model from a speaker’s enrollment data and uses it to recognize any new test speech. In this work, we propose a multi-model i-vector system for short speech lengths. We use an open database THUYG-20 for the analysis and development of short speech speaker verification and identification system. By using an optimum set of mel-frequency cepstrum coefficients (MFCC) based features we are able to achieve an equal error rate (EER) of 3.21% as compared to the previous benchmark score of EER 4.01% on the THUYG-20 database. Experiments are conducted for speech lengths as short as 0.25 s and the results are presented. The proposed method shows improvement as compared to the current i-vector based approach for shorter speech lengths. We are able to achieve improvement of around 28% even for 0.25 s speech samples. We also prepared and tested the proposed approach on our own database with 2500 speech recordings in English language consisting of actual short speech commands used in any voice interactive system.     

Dissociating automatic associations: Comparing two implicit measurements of race bias

Weak correspondence across different implicit bias tasks may arise from the contribution of unique forms of automatic and controlled processes to response behavior. Here, we examined the correspondence between estimates of automatic and controlled processing derived from two sequential priming tasks with identical structure and timing designed to separately measure stereotypic (Weapons Identification Task; WIT) and evaluative (Affective Priming Task; APT) associations. Across two studies using predominantly White samples, three consistent patterns emerged in the data: (a) stereotypic bias was stronger for Black targets, whereas evaluative bias was stronger for White targets; (b) overall response accuracy bias correlated modestly across the two tasks; and (c) multinomial processing tree estimates of controlled processing corresponded much more strongly than estimates of automatic processing. These findings support models positing distinct learning and memory systems for different forms of race bias, and suggest that these differing forms contribute to estimates of automatic associations.     

Out on a limb? On multiple cognitive systems within the octopus nervous system

The idea that there can be only one cognitive system within any single given cognitive organism is an established albeit implicit one within cognitive science and related studies of the mind. The firm foothold of this notion is due largely to the immense corpus of empirical evidence for the correlation of a high level of cognitive sophistication with a centralized nervous system. However, it must be pointed out that these findings are sourced in large part from studies on vertebrates. This paper presents a potential counterexample to the notion that only one cognitive system can be realized within any single genuine cognitive organism. This counterexample is the octopus, an invertebrate with what initially appears to be a paradoxical combination of vertebrate-like cognitive and behavioral capacities and a functionally decentralized nervous system. The extensive relegation of sensorimotor processing and control responsibilities to the peripheral nervous system which controls the arms of the octopus raises principled reasons to believe that the octopus is an organism that may house multiple independent cognitive systems.     

Generation and evaluation of user tailored responses in multimodal dialogue

When people engage in conversation, they tailor their utterances to their conversational partners, whether these partners are other humans or computational systems. This tailoring, or adaptation to the partner takes place in all facets of human language use, and is based on a mental model or a user model of the conversational partner. Such adaptation has been shown to improve listeners’ comprehension, their satisfaction with an interactive system, the efficiency with which they execute conversational tasks, and the likelihood of achieving higher level goals such as changing the listener’s beliefs and attitudes. We focus on one aspect of adaptation, namely the tailoring of the content of dialogue system utterances for the higher level processes of persuasion, argumentation and advice-giving. Our hypothesis is that algorithms that adapt content for these processes, according to a user model, will improve the usability, efficiency, and effectiveness of dialogue systems. We describe a multimodal dialogue system and algorithms for adaptive content selection based on multi-attribute decision theory. We demonstrate experimentally the improved efficacy of system responses through the use of user models to both tailor the content of system utterances and to manipulate their conciseness.     

Using an integrated simulation environment for the risk based design of advanced driver assistance systems

When designing new Advanced Driver Assistance Systems (ADAS), the existing guidelines like the RESPONSE 3 Code of Practice imply intense testing of the system prior series production. Within the EU-FP7-Project ISi-PADAS a new methodology of risk based design has been established which includes an integrated simulation platform. This integration enables intense testing of new prototypes in an accelerated way and therefore enhances the design process while directly combining the simulation results with a risk matrix.This paper describes the new methodology while focussing on the software framework and the procedure of risk assessment in detail. It shows the general approach as well as the appropriate steps taken for an ADAS prototype developed in the context of ISi-PADAS.     

Response interference under near-concurrent presentation of safety and non-safety information

The growing advent of Connected Vehicle Systems (CVS) is changing the information environment within the vehicle cab. As vehicles add the capability to receive data from the road infrastructure or share data with other vehicles, there is an expanding amount of safety and non-safety information from these systems with which the driver must contend. The information processing demands for the driver may become more complex, especially under conditions that present multiple information signals to the driver at the same time. To manage this complexity, the design of CVS will need to incorporate information management functions to prioritize information so as to be compatible with the processing capacity of the driver. The objective of this research was to examine the potential interference effect of non-safety critical information on driver responses to near concurrent, critical safety warnings. The study design was based on theory and evidence that there is a “bottle neck” in a human’s central processing of information, such that the processing of early signals (S1) in the environment may delay the response to a later signal (S2) until such time as the response to the first stimulus has been initiated. The results of the study suggest that there is the possibility that near concurrent presentation of safety and non-safety critical information may generate interference effects, but the combination of signal parameters (modality, timing) that are most likely to create this interference may be infrequent in most real world conditions. Future research should focus on the specific parameters that increase the probability of such interference. Such research would then provide a better estimate of the potential frequency of this interference and also guide the formulation of design guidelines to minimize this potential.     

Beyond Consistency: Contextual Dependency of Language Style in Monolog and Conversation

Language is highly dynamic: It unfolds over time, and we can use it to achieve a wide variety of communicative goals, from telling a story to trying to persuade another person. One aspect of language that has gained increasing popularity among researchers in the last several decades is the individual language style (LS) represented by an individual’s use of function words (e.g., pronouns, articles). Previous approaches to LS mostly focus on LS of one individual in isolation, paying less attention to the fact that language emerges from interaction with others. The aim of this paper is twofold: First, we integrate LS into a dynamical theoretical framework and present an innovative methodological approach. Second, this paper aims to address how interactive conversation—as an aspect of the communicative setting—changes an individual’s LS. We use recurrence quantification analysis to look at structure in patterns of LS in monologs and conversations of 118 participants. Our results showed that LS significantly differs from monolog to conversation, and post hoc analyses further revealed that the change in LS is greater for conflict than for friendly conversations. The difference between monologs and conversations is reflected more strongly in the dynamics (i.e., structure and complexity) of LS than the proportion of function words used. Theoretical implications and directions for future studies are discussed.