Cognitive dissonance | EPFL Graph Search

In the field of psychology, cognitive dissonance is the perception of contradictory information and the mental toll of it. Relevant items of information include a person's actions, feelings, ideas, beliefs, values, and things in the environment. Cognitive dissonance is typically experienced as psychological stress when persons participate in an action that goes against one or more of those things. According to this theory, when two actions or ideas are not psychologically consistent with each other, people do all in their power to change them until they become consistent. The discomfort is triggered by the person's belief clashing with new information perceived, wherein the individual tries to find a way to resolve the contradiction to reduce their discomfort. In When Prophecy Fails: A Social and Psychological Study of a Modern Group That Predicted the Destruction of the World (1956) and A Theory of Cogniti

Correlation-aware Resource Allocation in Multi-Cell Networks

Dorna Bandari, Pascal Frossard

We propose a cross-layer strategy for resource allocation between spatially correlated sources in the uplink of multi-cell FDMA networks. Our objective is to find the optimum power and channel allocation to the different sources, in order to minimize the maximum distortion achieved in decoding any source data in the network. This problem is NP-hard and finding the optimal solution is not computationally feasible. We propose a three-step algorithm to be performed separately in each cell, which finds cross-layer resource allocation in simple steps. This method separates the problem into inter-cell resource management, grouping of sources for joint decoding, and intra-cell channel assignment. For each of these steps we propose methods that satisfy different design constraints and analyze them by simulations. We show that, while using correlation in compression and joint decoding can achieve 25% distortion reduction over independent decoding, the improvement grows to 37% when correlation is also utilized in resource allocation. This significant distortion reduction motivates further work in correlation-aware resource allocation. Overall, our solution is able to achieve a 60% decrease in 5 percentile distortion compared to independent allocation methods.

Institute of Electrical and Electronics Engineers2012

Visuo-vestibular mechanisms of bodily self-consciousness

Joachim Jean-Marie Forget

Bodily self-consciousness is linked to multisensory integration and is particularly dependent on vestibular perception providing the brain with the main sensory cues about body motion and location in space. Vestibular and visual inputs are permanently balanced and integrated to encode the most optimal representation of the external world and of the observer in the central nervous system. Vection, an illusory self-motion experience induced only by visual stimuli, illustrates the fact that the visual and the vestibular systems share common neural underpinnings and a similar phenomenology. Optokinetic stimulation inducing vection and direct vestibular stimulation induce whole-body motion sensations that can be used to explore multisensory interactions. A failure in visuo-vestibular integration, artificially induced by the methods of cognitive psychology or in pathological conditions, has also been reported to altered own body perception and bodily self-consciousness. The respective contributions of the vestibular and visual systems to bodily self-consciousness amongst other polymodal sensory mechanisms, and the neural correlates of visuo-vestibular convergence, should be better understood. We first performed a neuroimaging study of brain regions where optokinetic and vestibular stimuli converge, using 7T functional magnetic resonance imaging in individual subjects. We identified three main regions of convergence: (1) the depth of supramarginal gyrus or retroinsular cortex, (2) the surface of supramarginal gyrus at the temporo-parietal junction, (3) and the posterior part of middle temporal gyrus and superior temporal sulcus. Then, we aimed to induce the embodiment of an external fake rubber hand through visuo-tactile conflict - the so-called rubber hand illusion paradigm, and studied how this integration is modulated by vection. Subjects experiencing vection in the direction of the rubber hand mislocalised the position of their real hand towards the rubber hand indicating that visuo-vestibular stimuli can enhance visuo-tactile integration. We also investigated if visuo-proprioceptive and tactile integration in peripersonal space could be dynamically updated based on the congruency of visual and proprioceptive feedback. A pair of rubber hands or feet provided visual feedback. Fake and real limbs were crossed or uncrossed. We showed that sensory cues were integrated in peripersonal space, dynamically reshaped but only for hands. Finally, we investigated a rare case of an illusory own body perception in an epileptic patient suffering from multiple daily disembodiments during seizures. Seizures were associated to a focal cortical microdysplasia juxtaposed to a developmental venous anomaly in the left angular gyrus, a brain region known to be important for visuo-vestibular integration and bodily self-consciousness. Our results characterize the inferior parietal lobule as a crucial structure in merging visual, vestibular, tactile and proprioceptive inputs, allowing the emergence of the global and unified experience of being âI.â Multisensory body representation can be reshaped transiently using visual and vestibular signals or in relation to a medical condition affecting the temporo-parietal junction. The integration of visual and vestibular signals, aims to adapt dynamically our internal representations to constant changes occurring in our environment.

EPFL2015

Truthful, Transparent and Fair Data Collection Mechanisms

Naman Goel

An important prerequisite for developing trustworthy artificial intelligence is high quality data. Crowdsourcing has emerged as a popular method of data collection in the past few years. However, there is always a concern about the quality of the data thus collected. This thesis addresses two major challenges in collecting high quality data from a crowd: 1) how to incentivize crowd workers to report accurate data; 2) how to ensure that the data collection mechanism is transparent and fair. We first propose two novel peer-consistency mechanisms for crowdsourcing: the Deep Bayesian Trust (DBT) mechanism and the Personalized Peer Truth Serum (PPTS). The DBT mechanism incentivizes workers to report accurate answers for objective questions with discrete ground truth answers. It is useful, for example, in collecting labels for supervised machine learning tasks. The mechanism ensures dominant uniform strategy incentive compatibility and fair rewards to the workers. The PPTS incentivizes workers to truthfully report their personal data (for example, body measurements). Since data is personal in nature, the tasks can not be shared between two workers. We show that when individuals report combinations of multiple personal data attributes, the correlation between them can be exploited to find peers and provide guarantees on the incentive compatibility of the mechanism. We next address the transparency issue of data collection. Smart contracts often rely on a trusted third party (oracle) to get correct information about real-world events. We show how peer-consistency mechanisms can be used to build decentralized, trustless and transparent data oracles on blockchain. We derive conditions under which a peer-consistency incentive mechanism can be used to acquire truthful information from an untrusted and self-interested crowd, even when the crowd has outside incentives to provide wrong information. We also show how to implement the peer-consistency mechanisms in Ethereum. We discuss various non-trivial issues that arise in implementing peer-consistency mechanisms in Ethereum, suggest several optimizations to reduce gas cost and provide empirical analysis. Finally, we address the problem of fair data collection from a crowd. Sharing economy platforms such as Airbnb and Uber face a major challenge in the form of peer-to-peer discrimination based on sensitive personal attributes such as race and gender. We show that how a peer-consistency incentive mechanism can be used to encourage users to go against common bias and provide a truthful rating about others, obtained through a more careful and deeper evaluation. In situations where an incentive mechanism canât be implemented, we show that a simple post-processing approach can also be used to correct bias in the reputation scores, while minimizing loss in the useful information provided by the scores. We also address the problem of fair and diverse data collection from a crowd under budget constraints. We propose a novel algorithm which maximizes the expected accuracy of the collected data, while ensuring that the errors satisfy desired notions of fairness w.r.t sensitive attributes.