The need to develop efficient therapies for neurodegenerative diseases is urgent, especially given the increasing percentages of the population living longer, with increasing chances of being afflicted with conditions like Parkinson's disease (PD). A promising curative approach toward PD and other neurodegenerative diseases is the transplantation of stem cells to halt and potentially reverse neuronal degeneration. However, stem cell therapy does not consistently lead to improvement for patients. Using remote stimulation to optogenetically activate transplanted cells, we attempted to improve behavioral outcomes of stem cell transplantation. We generated a neuronal precursor cell line expressing luminopsin 3 (LMO3), a luciferase-channelrhodopsin fusion protein, which responds to the luciferase substrate coelenterazine (CTZ) with emission of blue light that in turn activates the opsin. Neuronal precursor cells were injected bilaterally into the striatum of homozygous aphakia mice, which carry a spontaneous mutation leading to lack of dopaminergic neurons and symptoms of PD. Following transplantation, the cells were stimulated over a period of 10 days by intraventricular injections of CTZ. Mice receiving CTZ demonstrated significantly improved motor skills in a rotarod test compared to mice receiving vehicle. Thus, bioluminescent optogenetic stimulation of transplanted neuronal precursor cells shows promising effects in improving locomotor behavior in the aphakia PD mouse model and encourages further studies to elucidate the mechanisms and long-term outcomes of these beneficial effects.
Exploration in sparse reward reinforcement learning remains an open challenge. Many state-of-the-art methods use intrinsic motivation to complement the sparse extrinsic reward signal, giving the agent more opportunities to receive feedback during exploration. Commonly these signals are added as bonus rewards, which results in a mixture policy that neither conducts exploration nor task fulfillment resolutely. In this paper, we instead learn separate intrinsic and extrinsic task policies and schedule between these different drives to accelerate exploration and stabilize learning. Moreover, we introduce a new type of intrinsic reward denoted as successor feature control (SFC), which is general and not task-specific. It takes into account statistics over complete trajectories and thus differs from previous methods that only use local information to evaluate intrinsic motivation. We evaluate our proposed scheduled intrinsic drive (SID) agent using three different environments with pure visual inputs: VizDoom, DeepMind Lab and DeepMind Control Suite. The results show a substantially improved exploration efficiency with SFC and the hierarchical usage of the intrinsic drives. A video of our experimental results can be found at https://youtu.be/b0MbY3lUlEI.
Early stopping based on the validation set performance is a popular approach to find the right balance between under-and overfitting in the context of supervised learning. However, in reinforcement learning, even for supervised sub-problems such as world model learning, early stopping is not applicable as the dataset is continually evolving. As a solution, we propose a new general method that dynamically adjusts the update to data (UTD) ratio during training based on underand overfitting detection on a small subset of the continuously collected experience not used for training. We apply our method to DreamerV2, a state-of-the-art model-based reinforcement learning algorithm, and evaluate it on the DeepMind Control Suite and the Atari 100k benchmark. The results demonstrate that one can better balance under-and overestimation by adjusting the UTD ratio with our approach compared to the default setting in DreamerV2 and that it is competitive with an extensive hyperparameter search which is not feasible for many applications. Our method eliminates the need to set the UTD hyperparameter by hand and even leads to a higher robustness with regard to other learning-related hyperparameters further reducing the amount of necessary tuning.
Elongated protein‐based micro‐ and nanostructures are of great interest for a wide range of biomedical applications, where they can serve as a backbone for surface functionalization and as vehicles for drug delivery. Current production methods for protein constructs lack precise control of either shape and dimensions or render structures fixed to substrates. This work demonstrates production of recombinant spider silk nanowires suspended in solution, starting with liquid bridge induced assembly (LBIA) on a substrate, followed by release using ultrasonication, and concentration by centrifugation. The significance of this method lies in that it provides i) reproducability (standard deviation of length <13% and of diameter <38%), ii) scalability of fabrication, iii) compatibility with autoclavation with retained shape and function, iv) retention of bioactivity, and v) easy functionalization both pre‐ and post‐formation. This work demonstrates how altering the function and nanotopography of a surface by nanowire coating supports the attachment and growth of human mesenchymal stem cells (hMSCs). Cell compatibility is further studied through integration of nanowires during aggregate formation of hMSCs and the breast cancer cell line MCF7. The herein‐presented industrial‐compatible process enables silk nanowires for use as functionalizing agents in a variety of cell culture applications and medical research.
Real-time object detection in videos using lightweight hardware is a crucial component of many robotic tasks. Detectors using different modalities and with varying computational complexities offer different trade-offs. One option is to have a very lightweight model that can predict from all modalities at once for each frame. However, in some situations (e.g., in static scenes) it might be better to have a more complex but more accurate model and to extrapolate from previous predictions for the frames coming in at processing time. We formulate this task as a sequential decision making problem and use reinforcement learning (RL) to generate a policy that decides from the RGB input which detector out of a portfolio of different object detectors to take for the next prediction. The objective of the RL agent is to maximize the accuracy of the predictions per image. We evaluate the approach on the Waymo Open Dataset and show that it exceeds the performance of each single detector.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.