A recurrent neural network framework for flexible and adaptive decision making based on sequence learning

Sequential neuronal processing of number values, abstract decision, and action in the primate prefrontal cortex

Decision-making requires processing of sensory information, comparing the gathered evidence to make a judgment, and performing the action to communicate it. How neuronal representations transform during this cascade of representations remains a matter of debate. Here, we studied the succession of neuronal representations in the primate prefrontal cortex (PFC). We trained monkeys to judge whether a pair of sequentially presented displays had the same number of items. We used a combination of single neuron and population-level analyses and discovered a sequential transformation of represented information with trial progression. While numerical values were initially represented with high precision and in conjunction with detailed information such as order, the decision was encoded in a low-dimensional subspace of neural activity. This decision encoding was invariant to both retrospective numerical values and prospective motor plans, representing only the binary judgment of "same number" versus "different number," thus facilitating the generalization of decisions to novel number pairs. We conclude that this transformation of neuronal codes within the prefrontal cortex supports cognitive flexibility and generalizability of decisions to new conditions.

Increasing comprehensiveness and reducing workload in a systematic review of complex interventions using automated machine learning

BACKGROUND

As part of our ongoing systematic review of complex interventions for the primary prevention of cardiovascular diseases, we have developed and evaluated automated machine-learning classifiers for title and abstract screening. The aim was to develop a high-performing algorithm comparable to human screening.

METHODS

We followed a three-phase process to develop and test an automated machine learning-based classifier for screening potential studies on interventions for primary prevention of cardiovascular disease. We labelled a total of 16,611 articles during the first phase of the project. In the second phase, we used the labelled articles to develop a machine learning-based classifier. After that, we examined the performance of the classifiers in correctly labelling the papers. We evaluated the performance of the five deep-learning models [i.e. parallel convolutional neural network ( CNN ), stacked CNN , parallel-stacked CNN , recurrent neural network ( RNN ) and CNN-RNN]. The models were evaluated using recall, precision and work saved over sampling at no less than 95% recall.

RESULTS

We labelled a total of 16,611 articles, of which 676 (4.0%) were tagged as 'relevant' and 15,935 (96%) were tagged as 'irrelevant'. The recall ranged from 51.9% to 96.6%. The precision ranged from 64.6% to 99.1%. The work saved over sampling ranged from 8.9% to as high as 92.1%. The best-performing model was parallel CNN , yielding a 96.4% recall, as well as 99.1% precision, and a potential workload reduction of 89.9%.

FUTURE WORK AND LIMITATIONS

We used words from the title and the abstract only. More work needs to be done to look into possible changes in performance, such as adding features such as full document text. The approach might also not be able to be used for other complex systematic reviews on different topics.

CONCLUSION

Our study shows that machine learning has the potential to significantly aid the labour-intensive screening of abstracts in systematic reviews of complex interventions. Future research should concentrate on enhancing the classifier system and determining how it can be integrated into the systematic review workflow.

FUNDING

This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in Health Technology Assessment. See the NIHR Journals Library website for further project information.

An Evaluation of 3D-Printed Materials' Structural Properties Using Active Infrared Thermography and Deep Neural Networks Trained on the Numerical Data

This article describes an approach to evaluating the structural properties of samples manufactured through 3D printing via active infrared thermography. The mentioned technique was used to test the PETG sample, using halogen lamps as an excitation source. First, a simplified, general numerical model of the phenomenon was prepared; then, the obtained data were used in a process of the deep neural network training. Finally, the network trained in this manner was used for the material evaluation on the basis of the original experimental data. The described methodology allows for the automated assessment of the structural state of 3D-printed materials. The usage of a generalized model is an innovative method that allows for greater product assessment flexibility.

Latent characteristics and neural manifold of brain functional network under acupuncture

Acupuncture can regulate the cognition of brain system, and different manipulations are the keys of realizing the curative effect of acupuncture on human body. Therefore, it is crucial to distinguish and monitor the different acupuncture manipulations automatically. In this brief, in order to enhance the robustness of electroencephalogram (EEG) detection against noise and interference, we propose an acupuncture manipulation detecting framework based on supervised ISOMAP and recurrent neural network (RNN). Primarily, the low-dimensional embedding neural manifold of brain dynamical functional network is extracted via the reconstructed geodetic distance. It is found that there exhibits stronger acupuncture-specific reconfiguration of brain network. Besides, we show that the distance travel along this manifold correlates strongly with changes of acupuncture manipulations. The low-dimensional brain topological structure of all subjects shows crescent-like feature when acupuncturing at Zusanli acupoints, and fixed-points are varying under diverse manipulation methods. Moreover, Takagi-Sugeno-Kang (TSK) classifier is adopted to identify acupuncture manipulations according to the nonlinear characteristics of neural manifolds. Compared with different classifier, TSK can further improve the accuracy of manipulation identification at 96.71%. The results demonstrate the effectiveness of our model in detecting the acupuncture manipulations, which may provide neural biomarkers for acupuncture physicians.

Category learning in a recurrent neural network with reinforcement learning

It is known that humans and animals can learn and utilize category information quickly and efficiently to adapt to changing environments, and several brain areas are involved in learning and encoding category information. However, it is unclear that how the brain system learns and forms categorical representations from the view of neural circuits. In order to investigate this issue from the network level, we combine a recurrent neural network with reinforcement learning to construct a deep reinforcement learning model to demonstrate how the category is learned and represented in the network. The model consists of a policy network and a value network. The policy network is responsible for updating the policy to choose actions, while the value network is responsible for evaluating the action to predict rewards. The agent learns dynamically through the information interaction between the policy network and the value network. This model was trained to learn six stimulus-stimulus associative chains in a sequential paired-association task that was learned by the monkey. The simulated results demonstrated that our model was able to learn the stimulus-stimulus associative chains, and successfully reproduced the similar behavior of the monkey performing the same task. Two types of neurons were found in this model: one type primarily encoded identity information about individual stimuli; the other type mainly encoded category information of associated stimuli in one chain. The two types of activity-patterns were also observed in the primate prefrontal cortex after the monkey learned the same task. Furthermore, the ability of these two types of neurons to encode stimulus or category information was enhanced during this model was learning the task. Our results suggest that the neurons in the recurrent neural network have the ability to form categorical representations through deep reinforcement learning during learning stimulus-stimulus associations. It might provide a new approach for understanding neuronal mechanisms underlying how the prefrontal cortex learns and encodes category information.

Gated recurrence enables simple and accurate sequence prediction in stochastic, changing, and structured environments

From decision making to perception to language, predicting what is coming next is crucial. It is also challenging in stochastic, changing, and structured environments; yet the brain makes accurate predictions in many situations. What computational architecture could enable this feat? Bayesian inference makes optimal predictions but is prohibitively difficult to compute. Here, we show that a specific recurrent neural network architecture enables simple and accurate solutions in several environments. This architecture relies on three mechanisms: gating, lateral connections, and recurrent weight training. Like the optimal solution and the human brain, such networks develop internal representations of their changing environment (including estimates of the environment’s latent variables and the precision of these estimates), leverage multiple levels of latent structure, and adapt their effective learning rate to changes without changing their connection weights. Being ubiquitous in the brain, gated recurrence could therefore serve as a generic building block to predict in real-life environments.