This study presents the results of a series of simulation experiments that evaluate and compare four different manifold alignment methods under the influence of noise. The data was created by simulating the dynamics of two slightly different double pendulums in three-dimensional space. The method of semi-supervised feature-level manifold alignment using global distance resulted in the most convincing visualisations. However, the semi-supervised feature-level local alignment methods resulted in smaller alignment errors. These local alignment methods were also more robust to noise and faster than the other methods.
Structural alignment involves finding equivalences between sequential positions in two proteins. As such, it is similar to sequence alignment. However, in structural alignment the equivalences are not found by comparing two strings of characters but rather by optimally superimposing two structures and finding the regions of closest overlap in three-dimensions (figure 1). Structural alignment is becoming increasingly important as the number of known protein structures increases exponentially. Currently, there are more than 5000 structures in the Protein Data Bank (exactly, 5208 as of September 1995). Structural alignment is also very important because it is usually thought of as providing a standard or target for sequence alignment. That is, one will be a long way towards achieving accurate sequence alignment if one can align two homologous but highly diverged proteins (say, with low percent identity of-15 %) on the basis of sequence as well as on the basis of structure.
Methods from machine learning have successfully been used to improve the performance of control systems in cases when accurate models of the system or the environment are not available. These methods require the use of data generated from physical trials. Transfer Learning (TL) allows for this data to come from a different, similar system. The goal of this work is to understand in which cases a simple, alignment-based transfer of data is beneficial. A scalar, linear, time invariant(LTI) transformation is applied to the output from a source system to align with the output from a target system. In a theoretic study, we have already shown that for linear, single-input, single-output systems, the upper bound of the transformation error depends on the dynamic properties of the source and target system, and is small for systems with similar response times. We now consider two nonlinear, unicycle robots. Based on our previous work, we derive analytic error bounds for the linearized robot models. We then provide simulations of the nonlinear robot models and experiments with a Pioneer 3-AT robot that confirm the theoretical findings. As a result, key characteristics of alignment based transfer learning observed in our theoretic study prove to be also true for real, nonlinear unicycle robots.
There has been a sense that as the capabilities of artificial intelligence has expanded at a rapid pace in the past few years that we need to step back and think of the philosophical and ethical side of AI. This is especially so when we have such a patchy understanding of how seemingly straightforward goals might be carried out by an AI. For instance, requesting that an AI eradicate cancer could prompt it to kill all humans, thus achieving its ultimate goal but probably not in the way we'd desire. Researchers from the Georgia Institute of Technology believe that robots can learn sufficient ethics, even if it's not hardwired into them by using an approach they're calling Quixote. The approach, which was documented in a recent paper, uses value alignment, with the robots trained using stories to understand right from wrong.
Obtaining an accurate multiple alignment of protein sequences is a difficult computational problem for which many heuristic techniques sacrifice optimality to achieve reasonable running times. The most commonly used heuristic is progressive alignment, which merges sequences into a multiple alignment by pairwise comparisons along the nodes of a guide tree. To improve accuracy, consistency-based methods take advantage of conservation across many sequences to provide a stronger signal for pairwise comparisons. In this paper, we introduce the concept of probabilistic consistency for multiple sequence alignments.