A high-throughput screening approach to discovering

نویسندگان

  • Nicolas Pinto
  • David Doukhan
  • James J. DiCarlo
  • David D. Cox
چکیده

While many models of biological object recognition share a common set of ‘‘broad-stroke’’ properties, the performance of any one model depends strongly on the choice of parameters in a particular instantiation of that model—e.g., the number of units per layer, the size of pooling kernels, exponents in normalization operations, etc. Since the number of such parameters (explicit or implicit) is typically large and the computational cost of evaluating one particular parameter set is high, the space of possible model instantiations goes largely unexplored. Thus, when a model fails to approach the abilities of biological visual systems, we are left uncertain whether this failure is because we are missing a fundamental idea or because the correct ‘‘parts’’ have not been tuned correctly, assembled at sufficient scale, or provided with enough training. Here, we present a high-throughput approach to the exploration of such parameter sets, leveraging recent advances in stream processing hardware (high-end NVIDIA graphic cards and the PlayStation 3’s IBM Cell Processor). In analogy to highthroughput screening approaches in molecular biology and genetics, we explored thousands of potential network architectures and parameter instantiations, screening those that show promising object recognition performance for further analysis. We show that this approach can yield significant, reproducible gains in performance across an array of basic object recognition tasks, consistently outperforming a variety of state-of-the-art purpose-built vision systems from the literature. As the scale of available computational power continues to expand, we argue that this approach has the potential to greatly accelerate progress in both artificial vision and our understanding of the computational underpinning of biological vision. Citation: Pinto N, Doukhan D, DiCarlo JJ, Cox DD (2009) A High-Throughput Screening Approach to Discovering Good Forms of Biologically Inspired Visual Representation. PLoS Comput Biol 5(11): e1000579. doi:10.1371/journal.pcbi.1000579 Editor: Karl J. Friston, University College London, United Kingdom Received June 11, 2009; Accepted October 26, 2009; Published November 26, 2009 Copyright: 2009 Pinto et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This study was funded in part by The National Institutes of Health (NEI R01EY014970), The McKnight Endowment for Neuroscience, Dr. Gerald Burnett and Marjorie Burnett, and The Rowland Institute of Harvard. Hardware support was generously provided by the NVIDIA Corporation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] (NP); [email protected] (DDC)

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Nootropic Medicinal Plants; Evaluating Potent Formulation By Novelestic High throughput Pharmacological Screening (HTPS) Method

The principle of this method was to screen the pharmacological activity of six prepared polyphyto formulations by using high throughput screening method for their nootropic action. The study was performed in three stages using one, two and three animals, respectively in a group. Test formulations were given p.o daily at the dose of 50 and 100 mg/kg body weight. The test formulations were compar...

متن کامل

Add-on for High Throughput Screening in Material Discovery for Organic Electronics: “Tagging” Molecules to Address the Device Considerations

This work reflects the worth of intelligent modeling in controlling the nanostructure morphology in manufacturing organic bulk heterojunction (BHJ) solar cells. It suggests the idea of screening the pool of material design possibilities inspired by machine learning. To fulfill this goal, a set of experimental data on a BHJ solar cell with a donor structure of diketopyrrolopyrrole (DDP) and ...

متن کامل

SbacHTS: Spatial background noise correction for High-Throughput RNAi Screening

MOTIVATION High-throughput cell-based phenotypic screening has become an increasingly important technology for discovering new drug targets and assigning gene functions. Such experiments use hundreds of 96-well or 384-well plates, to cover whole-genome RNAi collections and/or chemical compound files, and often collect measurements that are sensitive to spatial background noise whose patterns ca...

متن کامل

Assessment of "drug-likeness" of a small library of natural products using chemoinformatics

Even though natural products has an excellent record as a source for new drugs, the advent of ultrahigh-throughput screening and large-scale combinatorial synthetic methods, has caused a decline in the use of natural products research in the pharmaceutical industry. This is due to the efficiency in generating and screening a high number of synthetic combinatorial compounds; whereas traditional ...

متن کامل

Assessment of "drug-likeness" of a small library of natural products using chemoinformatics

Even though natural products has an excellent record as a source for new drugs, the advent of ultrahigh-throughput screening and large-scale combinatorial synthetic methods, has caused a decline in the use of natural products research in the pharmaceutical industry. This is due to the efficiency in generating and screening a high number of synthetic combinatorial compounds; whereas traditional ...

متن کامل

Discovering reliable protein interactions from high-throughput experimental data using network topology

OBJECTIVE Current protein-protein interaction (PPI) detection via high-throughput experimental methods, such as yeast-two-hybrid has been reported to be highly erroneous, leading to potentially costly spurious discoveries. This work introduces a novel measure called IRAP, i.e. "interaction reliability by alternative path", for assessing the reliability of protein interactions based on the under...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2009