Newest and Lowest Rated Papers

Publication Date

Newest Oldest Clear

Average Rating

Highest Lowest Clear

Deep Hierarchies in the Primate Visual Cortex: What Can We Learn for Computer Vision?

N. Kruger, P. Janssen, S. Kalkan, M. Lappe, A. Leonardis, J. Piater, A. J. Rodriguez-Sanchez, L. Wiskott

IEEE Transactions on Pattern Analysis and Machine Intelligence·2013·303 citations

No abstract available

No ratings yet

View paper →

Why vision is not both hierarchical and feedforward

Michael H. Herzog, Aaron M. Clarke

Frontiers in Computational Neuroscience·2014·30 citations

No abstract available

No ratings yet

View paper →

The CRE1 carbon catabolite repressor of the fungus Trichoderma reesei: a master regulator of carbon assimilation

Thomas Portnoy, Antoine Margeot, Rita Linke, Lea Atanasova, Erzsébet Fekete, Erzsébet Sándor, Lukas Hartl, Levente Karaffa, Irina S Druzhinina, Bernhard Seiboth, Stéphane Le Crom, Christian P Kubicek

BMC Genomics·2011·180 citations

Abstract Background The identification and characterization of the transcriptional regulatory networks governing the physiology and adaptation of microbial cells is a key step in understanding their behaviour. One such wide-domain regulatory circuit, essential to all cells, is carbon catabolite repression (CCR): it allows the cell to prefer some carbon sources, whose assimilation is of high nutritional value, over less profitable ones. In lower multicellular fungi, the C2H2 zinc finger CreA/CRE1 protein has been shown to act as the transcriptional repressor in this process. However, the complete list of its gene targets is not known. Results Here, we deciphered the CRE1 regulatory range in the model cellulose and hemicellulose-degrading fungus Trichoderma reesei (anamorph of Hypocrea jecorina) by profiling transcription in a wild-type and a delta-cre1 mutant strain on glucose at constant growth rates known to repress and de-repress CCR-affected genes. Analysis of genome-wide microarrays reveals 2.8% of transcripts whose expression was regulated in at least one of the four experimental conditions: 47.3% of which were repressed by CRE1, whereas 29.0% were actually induced by CRE1, and 17.2% only affected by the growth rate but CRE1 independent. Among CRE1 repressed transcripts, genes encoding unknown proteins and transport proteins were overrepresented. In addition, we found CRE1-repression of nitrogenous substances uptake, components of chromatin remodeling and the transcriptional mediator complex, as well as developmental processes. Conclusions Our study provides the first global insight into the molecular physiological response of a multicellular fungus to carbon catabolite regulation and identifies several not yet known targets in a growth-controlled environment.

No ratings yet

View paper →

Uma proposta didática de resolução de problemas na matemática : escrever para entender, entender para resolver

Kleber Xavier Feitosa

No abstract available

No ratings yet

View paper →

Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networks

Alex Quach, Makram Chahine, Alexander Amini, Ramin Hasani, Daniela Rus

arXiv·2024

Simulators are powerful tools for autonomous robot learning as they offer scalable data generation, flexible design, and optimization of trajectories. However, transferring behavior learned from simulation data into the real world proves to be difficult, usually mitigated with compute-heavy domain randomization methods or further model fine-tuning. We present a method to improve generalization and robustness to distribution shifts in sim-to-real visual quadrotor navigation tasks. To this end, we first build a simulator by integrating Gaussian Splatting with quadrotor flight dynamics, and then, train robust navigation policies using Liquid neural networks. In this way, we obtain a full-stack imitation learning protocol that combines advances in 3D Gaussian splatting radiance field rendering, crafty programming of expert demonstration training data, and the task understanding capabilities of Liquid networks. Through a series of quantitative flight tests, we demonstrate the robust transfer of navigation skills learned in a single simulation scene directly to the real world. We further show the ability to maintain performance beyond the training environment under drastic distribution and physical environment changes. Our learned Liquid policies, trained on single target manoeuvres curated from a photorealistic simulated indoor flight only, generalize to multi-step hikes onboard a real hardware platform outdoors.

No ratings yet

View paper →

Robust flight navigation out of distribution with liquid neural networks

Makram Chahine, Ramin Hasani, Patrick Kao, Aaron Ray, Ryan Shubert, Mathias Lechner, Alexander Amini, Daniela Rus

Science Robotics·2023·56 citations

Autonomous robots can learn to perform visual navigation tasks from offline human demonstrations and generalize well to online and unseen scenarios within the same environment they have been trained on. It is challenging for these agents to take a step further and robustly generalize to new environments with drastic scenery changes that they have never encountered. Here, we present a method to create robust flight navigation agents that successfully perform vision-based fly-to-target tasks beyond their training environment under drastic distribution shifts. To this end, we designed an imitation learning framework using liquid neural networks, a brain-inspired class of continuous-time neural models that are causal and adapt to changing conditions. We observed that liquid agents learn to distill the task they are given from visual inputs and drop irrelevant features. Thus, their learned navigation skills transferred to new environments. When compared with several other state-of-the-art deep agents, experiments showed that this level of robustness in decision-making is exclusive to liquid networks, both in their differential equation and closed-form representations.

No ratings yet

View paper →

Neural Circuit Policies Imposing Visual Perceptual Autonomy

Waleed Razzaq, Mo Hongwei

Neural Processing Letters·2023·2 citations

We presented a hybrid deep learning architecture by combining biological brain-inspired Neural Circuit Policies with convolutional neural architecture. This allows the agent to learn key coherent features from various environments, express generalizability and interpret dynamics. We found that a combination of 12 inter-neurons and 8 command neurons can effectively perform high-stakes decision-making tasks such as autonomous vehicles. Our architecture not only has 305 times fewer trainable parameters than competing approaches but shows superior performance, robustness, and noise resiliency in famous autonomous vehicle simulations i.e. OpenAI-CarRacing and Udacity Simulator.

No ratings yet

View paper →

Ritualization of Affection and Respect: Two Principles of Confucian Ritual

Jaesang Jung

Religions·2019·2 citations

Confucian rituals have constituted the foundation of religious practice in the traditional societies of East Asia. Paying attention to the Confucian ritual, this article explores the way Confucianism constructs its symbolic system based on people’s natural feelings, particularly in the case of three-year mourning. It intends to show how the two feelings of “affection for the family” (chinchin/qinqin, 親親) and “respect for the honorable” (chonjon/zunzun, 尊尊) are ritualized in Confucian rites, and to illuminate the religious and social dimensions of Confucianism in premodern Korea by analyzing a seventeenth-century controversy over royal mourning from the perspective of these two principles.

No ratings yet

View paper →