We will in this paper identify some of the central problems of current techniques for rule extraction from recurrent neural networks (RNN-RE). Then we will raise the expectations of future RNN-RE techniques considerably and through this, hopefully guide the research towards a common goal. Some preliminary results based on work in line with these goals, will also be presented.

This project uses ideas from greedy transition based parsing to build neural network models that can jointly learn to parse sentences and use those parses to guide semantic composition. The model is used for sentence encoding for tasks like Sentiment classification and Entailment. The performance is evaluated on Stanford Sentiment Treebank(SST) and Stanford Natural Language Inference (SNLI) cor...

Self-organized criticality emerged in neural activity is one of the key concepts to describe the formation and the function of developing neuronal networks. The relationship between critical dynamics and neural development is both theoretically and experimentally appealing. However, whereas it is well-known that cortical networks exhibit a rich repertoire of activity patterns at different stage...

Since its introduction, multivariate pattern analysis (MVPA), or ‘neural decoding’, has transformed the field of cognitive neuroscience. Underlying its influence is a crucial inference, which we call the decoder’s dictum: if information can be decoded from patterns of neural activity, then this provides strong evidence about what information those patterns represent. Although the dictum is a wi...

Neural prosthesis has been successfully applied to patients with motional or sensory disabilities for clinical purpose. To enhance the performance of the neural prosthetic device, the electrodes for the biosignal recording or electrical stimulation should be located in closer proximity to target neurons than they are now. Instead of revising the prior implanting surgery to improve the electrica...

Learning from rewards and punishments is essential to survival and facilitates flexible human behavior. It is widely appreciated that multiple cognitive and reinforcement learning systems contribute to decision-making, but the nature of their interactions is elusive. Here, we leverage methods for extracting trial-by-trial indices of reinforcement learning (RL) and working memory (WM) in human e...

Understanding the principles of information processing in neural circuits requires systematic characterization of the participating cell types and their connections, and the ability to measure and perturb their activity. Genetic approaches promise to bring experimental access to complex neural systems, including genetic stalwarts such as the fly and mouse, but also to nongenetic systems such as...

Neural prostheses aim to improve the quality of life of severely disabled patients by translating neural activity into control signals for guiding prosthetic devices or computer cursors. We recently demonstrated that plan activity from premotor cortex, which specifies the endpoint of the upcoming arm movement, can be used to swiftly and accurately guide computer cursors to the desired target lo...

Probabilistic programming languages (PPLs) are a powerful modeling tool, ableto represent any computable probability distribution. Unfortunately, probabilisticprogram inference is often intractable, and existing PPLs mostly rely on expensive,approximate sampling-based methods. To alleviate this problem, one could tryto learn from past inferences, so that future inferences run fa...

A multi layered feed forward neural network model is being developed for the prediction of end blow oxygen in the LD converter using a two step process. In the first step intermediate stopping temperature is being predicted and using this as an input the end blow oxygen is predicted. In both the cases two hidden layers had given the best results compared to the single layer neural network. Inte...