Bar-Ilan University

About: Bar-Ilan University is a education organization based out in Ramat Gan, Israel. It is known for research contribution in the topics: Population & Poison control. The organization has 12835 authors who have published 34964 publications receiving 995648 citations. The organization is also known as: Bar Ilan University & BIU.

The Use of Redox Mediators for Enhancing Utilization of Li2S Cathodes for Advanced Li–S Battery Systems

Abstract: The development of Li2S electrodes is a crucial step toward industrial manufacturing of Li-S batteries, a promising alternative to Li-ion batteries due to their projected two times higher specific capacity. However, the high voltages needed to activate Li2S electrodes, and the consequent electrolyte solution degradation, represent the main challenge. We present a novel concept that could make feasible the widespread application of Li2S electrodes for Li-S cell assembly. In this concept, the addition of redox mediators as additives to the standard electrolyte solution allows us to recover most of Li2S theoretical capacity in the activation cycle at potentials as low as 2.9 VLi, substantially lower than the typical potentials >4 VLi needed with standard electrolyte solution. Those novel additives permit us to preserve the electrolyte solution from being degraded, allowing us to achieve capacity as high as 500 mAhg(-1)Li2S after 150 cycles with no major structural optimization of the electrodes.

Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection

Abstract: The ability to control for the kinds of information encoded in neural representation has a variety of use cases, especially in light of the challenge of interpreting these models. We present Iterative Null-space Projection (INLP), a novel method for removing information from neural representations. Our method is based on repeated training of linear classifiers that predict a certain property we aim to remove, followed by projection of the representations on their null-space. By doing so, the classifiers become oblivious to that target property, making it hard to linearly separate the data according to it. While applicable for multiple uses, we evaluate our method on bias and fairness use-cases, and show that our method is able to mitigate bias in word embeddings, as well as to increase fairness in a setting of multi-class classification.

The influence of stimulus deviance and novelty on the P300 and novelty P3.

Abstract: This study examined the relationship between ERP components elicited by deviant stimuli by disentangling the P300 and Novelty P3 components, using spatiotemporal principal components analysis and a dense electrode array. The three-tone paradigm was used and the pitch attributes of the tones were systematically manipulated so as to map the amplitude of the ERP components on the stimulus context. A comparison was made between the components elicited by events in the three-stimulus, classical oddball, and novelty oddball paradigms. Responses to deviant stimuli consisted of independent and dissociable ERP components in the 400-600-ms time range: A parietal component (P300) that was larger for targets than rare nontargets and was affected by the difficulty of discrimination, a fronto-central component (Novelty P3) that was larger for novel tones and for rare nontargets in the difficult discrimination condition, and an additional anterior negative component responded similarly to all types of deviant stimuli.

Scaling Web-based Acquisition of Entailment Relations.

Abstract: Paraphrase recognition is a critical step for natural language interpretation. Accordingly, many NLP applications would benefit from high coverage knowledge bases of paraphrases. However, the scalability of state-of-the-art paraphrase acquisition approaches is still limited. We present a fully unsupervised learning algorithm for Web-based extraction of entailment relations, an extended model of paraphrases. We focus on increased scalability and generality with respect to prior work, eventually aiming at a full scale knowledge base. Our current implementation of the algorithm takes as its input a verb lexicon and for each verb searches the Web for related syntactic entailment templates. Experiments show promising results with respect to the ultimate goal, achieving much better scalability than prior Web-based methods.

Activation of inhibition: Diminishing impulsive behavior by direct current stimulation over the inferior frontal gyrus

Abstract: A common feature of human existence is the ability to reverse decisions after they are made but before they are implemented. This cognitive control process, termed response inhibition, refers to the ability to inhibit an action once initiated and has been localized to the right inferior frontal gyrus (rIFG) based on functional imaging and brain lesion studies. Transcranial direct current stimulation (tDCS) is a brain stimulation technique that can facilitate as well as impair cortical function. To explore whether response inhibition can be improved through rIFG electrical stimulation, we administered focal tDCS before subjects performed the stop signal task (SST), which measures response inhibition. Notably, activation of the rIFG by unilateral anodal stimulation significantly improved response inhibition, relative to a sham condition, whereas the same tDCS protocol did not affect response time in the go trials of the SST and in a control task. Furthermore, the SST was not affected by tDCS at a control site, the right angular gyrus. Our results are the first demonstration of response inhibition improvement with brain stimulation over rIFG and further confirm the rIFG involvement in this task. Although this study was conducted in healthy subjects, present findings with anodal rIFG stimulation support the use of similar paradigms for the treatment of cognitive control impairments in pathological conditions.