A Linear Belief Function Approach to Portfolio Evaluation

arXiv.org Artificial Intelligence

By elaborating on the notion of linear belief functions (Dempster 1990; Liu 1996), we propose an elementary approach to knowledge representation for expert systems using linear belief functions. We show how to use basic matrices to represent market information and financial knowledge, including complete ignorance, statistical observations, subjective speculations, distributional assumptions, linear relations, and empirical asset pricing models. We then appeal to Dempster's rule of combination to integrate the knowledge for assessing an overall belief of portfolio performance, and updating the belief by incorporating additional information. We use an example of three gold stocks to illustrate the approach.

A Deformable Interface for Human Touch Recognition using Stretchable Carbon Nanotube Dielectric Elastomer Sensors and Deep Neural Networks

arXiv.org Machine Learning

User interfaces provide an interactive window between physical and virtual environments. A new concept in the field of human-computer interaction is a soft user interface; a compliant surface that facilitates touch interaction through deformation. Despite the potential of these interfaces, they currently lack a signal processing framework that can efficiently extract information from their deformation. Here we present OrbTouch, a device that uses statistical learning algorithms, based on convolutional neural networks, to map deformations from human touch to categorical labels (i.e., gestures) and touch location using stretchable capacitor signals as inputs. We demonstrate this approach by using the device to control the popular game Tetris. OrbTouch provides a modular, robust framework to interpret deformation in soft media, laying a foundation for new modes of human computer interaction through shape changing solids.

QSPR and QSAR Models Derived with CODESSA Multipurpose Statistical Analysis Software

AAAI Conferences

An overview on the development of QSPR/QSAR equations using various descriptor mining techniques and multilinear regression analysis in the framework of program CODESSA (Comprehensive Descriptors for Structural and Statistical Analysis) is given. The description of the methodologies applied in CODESSA is followed by the presentation of the QSAR and QSPR models derived for eighteen molecular activities and properties. The properties cover single molecular species, interactions between different molecular species, properties of surfactants, complex properties and properties of polymers.

Retrosynthetic reaction prediction using neural sequence-to-sequence models

arXiv.org Machine Learning

We describe a fully data driven model that learns to perform a retrosynthetic reaction prediction task, which is treated as a sequence-to-sequence mapping problem. The end-to-end trained model has an encoder-decoder architecture that consists of two recurrent neural networks, which has previously shown great success in solving other sequence-to-sequence prediction tasks such as machine translation. The model is trained on 50,000 experimental reaction examples from the United States patent literature, which span 10 broad reaction types that are commonly used by medicinal chemists. We find that our model performs comparably with a rule-based expert system baseline model, and also overcomes certain limitations associated with rule-based expert systems and with any machine learning approach that contains a rule-based expert system component. Our model provides an important first step towards solving the challenging problem of computational retrosynthetic analysis.

A QSAR - Bayesian neural network model to identify molecular properties causing eye irritation in cationic surfactants

AAAI Conferences

QSAR models are frequently used to investigate and predict the toxicological effects of chemicals. Building QSAR models of the eye irritation potential of cationic surfactants is difficult, as the mechanism of action of these surfactants is still not fully understood. This report describes a data driven QSAR model to predict Maximum Average Scores (MAS in accordance to Draize) for cationic surfactants from the calculated molecular properties Log P, Log CMC and molecular volume, and the surfactant concentration. We demonstrate that a Bayesian Neural Network, a statistical nonlinear regression approach that estimates the noise in the modelling data and error bars on the predictions, provided the most robust and accurate representation of the relationship between the MAS score and the molecular properties. A dataset of 20 in vivo rabbit eye irritation tests on 19 different cationic surfactants, obtained from historic in-house data and the scientific literature, was used to train the Bayesian neural network. The model was then used to simulate a large number of molecules to explore the relationship between MAS score and molecular properties. MAS vs. Log P showed bell shaped curve as expected. A higher concentration ( 20%) was required in order to elicit the eye irritancy response of molecules with a wide range of Log P. The simulated results were used to identify the range of molecular properties of cationic surfactants most likely to cause more than mild irritancy.