Hydrogen Production Using a Nickel Catalyst Combining Redox Activity and Pendent Base Effects.

With the mounting need for clean and renewable energy, catalysts for hydrogen production based on earth abundant elements are of great interest. Herein, we describe the synthesis, characterization, and catalytic activity of two nickel complexes based on the pyridinediimine ligand that possess basic nitrogen moieties of pyridine and imidazole that could potentially serve as pendent bases to enhance catalysis. Although these ligands have previously been reported to be complexed to some metal ions, they have not been applied to nickel. The nickel complex with the pendent pyridines was found to be the most active of the two, catalyzing proton reduction electrochemically with an overpotential of 490 mV. The appearance of a wave that preceded the Ni(I/0) redox couple in the presence of protons suggests that protonation of a dissociated pyridine was likely. Further evidence of this was provided with density functional theory calculations, and a mechanism of hydrogen production is proposed. Furthermore, in a light-driven system containing Ru(bpy)32+ and ascorbic acid, TON of 1400 were obtained.

Hybrid computing using a neural network with dynamic external memory.

Artificial neural networks are remarkably adept at sensory processing, sequence learning and reinforcement learning, but are limited in their ability to represent variables and data structures and to store data over long timescales, owing to the lack of an external memory. Here we introduce a machine learning model called a differentiable neural computer (DNC), which consists of a neural network that can read from and write to an external memory matrix, analogous to the random-access memory in a conventional computer. Like a conventional computer, it can use its memory to represent and manipulate complex data structures, but, like a neural network, it can learn to do so from data. When trained with supervised learning, we demonstrate that a DNC can successfully answer synthetic questions designed to emulate reasoning and inference problems in natural language. We show that it can learn tasks such as finding the shortest path between specified points and inferring the missing links in randomly generated graphs, and then generalize these tasks to specific graphs such as transport networks and family trees. When trained with reinforcement learning, a DNC can complete a moving blocks puzzle in which changing goals are specified by sequences of symbols. Taken together, our results demonstrate that DNCs have the capacity to solve complex, structured tasks that are inaccessible to neural networks without external read-write memory.

Human-level control through deep reinforcement learning.

The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past experience to new situations. Remarkably, humans and other animals seem to solve this problem through a harmonious combination of reinforcement learning and hierarchical sensory processing systems, the former evidenced by a wealth of neural data revealing notable parallels between the phasic signals emitted by dopaminergic neurons and temporal difference reinforcement learning algorithms. While reinforcement learning agents have achieved some successes in a variety of domains, their applicability has previously been limited to domains in which useful features can be handcrafted, or to domains with fully observed, low-dimensional state spaces. Here we use recent advances in training deep neural networks to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games. We demonstrate that the deep Q-network agent, receiving only the pixels and the game score as inputs, was able to surpass the performance of all previous algorithms and achieve a level comparable to that of a professional human games tester across a set of 49 games, using the same algorithm, network architecture and hyperparameters. This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.

MulTi-FAST: A Machinability Assessment of Functionally Graded Titanium Billets Produced from Multiple Alloy Powders.

Technological developments in the area of functionally graded multi-material manufacture are poised to disrupt the aerospace industry, providing the means for step-change improvements in performance through tailored component design. However, the challenges faced during the downstream processing, i.e., machining of such functionally graded multi-materials are unclear. In this study, the challenges involved when face-turning billets consisting of multiple alloys are assessed. To achieve this, a cylindrical billet consisting of Ti-64, Ti-6242, Ti-5553 and Beta C alloys was manufactured from powder feedstock using field-assisted sintering technique (FAST) and termed MulTi-FAST billets. A detailed study of the structural integrity during machining at the diffusion bond interfaces of multiple titanium alloy bond pairings in the MulTi-FAST billet was conducted. The machining forces were measured during face-turning to investigate the impact and behaviour of different alloy pairings during a continuous machining operation. The results showed the significant differences in force machining response, surface topography and the type of surface damage was dependent on the direction the titanium alloy graded pairings were machined in. In terms of subsurface microstructural damage, regardless of the machining direction, no critical damage was found in the vicinity of the bonded alloys. The findings provide an insight into the deformation characteristics and challenges faced in the machining of functionally graded components with multiple titanium alloys.

A novel connectionist system for unconstrained handwriting recognition.

Recognizing lines of unconstrained handwritten text is a challenging task. The difficulty of segmenting cursive or overlapping characters, combined with the need to exploit surrounding context, has led to low recognition rates for even the best current recognizers. Most recent progress in the field has been made either through improved preprocessing or through advances in language modeling. Relatively little work has been done on the basic recognition algorithms. Indeed, most systems rely on the same hidden Markov models that have been used for decades in speech and handwriting recognition, despite their well-known shortcomings. This paper proposes an alternative approach based on a novel type of recurrent neural network, specifically designed for sequence labeling tasks where the data is hard to segment and contains long-range bidirectional interdependencies. In experiments on two large unconstrained handwriting databases, our approach achieves word recognition accuracies of 79.7 percent on online data and 74.1 percent on offline data, significantly outperforming a state-of-the-art HMM-based system. In addition, we demonstrate the network's robustness to lexicon size, measure the individual influence of its hidden layers, and analyze its use of context. Last, we provide an in-depth discussion of the differences between the network and HMMs, suggesting reasons for the network's superior performance.

Framewise phoneme classification with bidirectional LSTM and other neural network architectures.

In this paper, we present bidirectional Long Short Term Memory (LSTM) networks, and a modified, full gradient version of the LSTM learning algorithm. We evaluate Bidirectional LSTM (BLSTM) and several other network architectures on the benchmark task of framewise phoneme classification, using the TIMIT database. Our main findings are that bidirectional networks outperform unidirectional ones, and Long Short Term Memory (LSTM) is much faster and also more accurate than both standard Recurrent Neural Nets (RNNs) and time-windowed Multilayer Perceptrons (MLPs). Our results support the view that contextual information is crucial to speech processing, and suggest that BLSTM is an effective architecture with which to exploit it.

Steps against recurrent stroke plus: patient transition program.

Stroke is a devastating health event that affects 800,000 people annually in the United States. Nearly 20% of strokes are recurrent strokes. Research shows that support after discharge from the hospital poststroke is frequently inadequate. The purpose of "Steps Against Recurrent Stroke (STARS) Plus: Patient Transition Program" was to design and deliver a program to facilitate optimal recovery for stroke survivors and prevent recurrent stroke. The program began at discharge from the hospital and continued through the first year of rehabilitation and recovery. Twelve hospitals participated; 261 patients enrolled, and contact was established with 193. Outcomes were gathered based on patient self-report of health status using the Short-Form Health Survey at 30, 90, 180, and 360 days. A dependent sample t test was completed comparing participants' 30- and 360-day follow-up scores. Results demonstrated an overall increase in subjective pain. A repeated multivariate analysis of variance was conducted to compare 30- and 360-day Short-Form Health Survey scores across age and subscales. Results revealed that those in the younger and older age groups reported poorer health outcomes. Findings demonstrate a reduction in rehospitalization after stroke, increased medication adherence, strong patient satisfaction, and significant differences in health-related outcome measures across age groups, suggesting that middle-aged stroke survivors experience better health outcomes than younger or older age groups. Future programs should consider targeting pain management in all ages and education targeted at younger and older age groups, because they reported poorer health outcomes. The findings from this program should contribute to the guidance and insight for others developing transitional interventions for stroke survivors.

Parameter-exploring policy gradients.

We present a model-free reinforcement learning method for partially observable Markov decision problems. Our method estimates a likelihood gradient by sampling directly in parameter space, which leads to lower variance gradient estimates than obtained by regular policy gradient methods. We show that for several complex control tasks, including robust standing with a humanoid robot, this method outperforms well-known algorithms from the fields of standard policy gradients, finite difference methods and population based heuristics. We also show that the improvement is largest when the parameter samples are drawn symmetrically. Lastly we analyse the importance of the individual components of our method by incrementally incorporating them into the other algorithms, and measuring the gain in performance after each step.