Unlocking Brain-Inspired Computer Vision

Nicolas Pinto

PhD candidate, Computational Neuroscience, Massachusetts Institute of Technology

Abstract

1. The study of biological vision and the creation of artificial vision systems are naturally intertwined—exploration of the neuronal substrates of visual processing provides clues and inspiration for artificial systems, and artificial systems, in turn, serve as important generators of new ideas and working hypotheses. However, while systems neuroscience has provided inspiration for some of the “broad-stroke” properties of the visual system, much is still unknown. Even for those qualitative properties that most biologically-inspired models share, experimental data currently provide little constraint on their key parameters. Consequently, it is difficult to truly evaluate a set of computational ideas, since the performance of a model depends strongly on its particular instantiation—the size of the pooling kernels, the number of units per layer, exponents in normalization operations, etc.
   

2. To pave a way forward, we have developed a high-throughput approach to more expansively explore the possible range of biologically-inspired models, including models of larger, more realistic scale, leveraging recent advances in commodity stream processing hardware—particularly, high-end NVIDIA GPUs.
   

3. In analogy to high-throughput screening approaches in molecular biology and genetics, we generated and trained thousands of potential network architectures and parameter instantiations, and “screened” the visual representations produced by these models using an object recognition task. From these candidate models, the most promising were selected for further analysis. We have shown 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, and that it can offer insight into which computational ideas are most important for achieving this performance.
   

4. In this talk, I'll also highlight how the application of flexible programming tools, such as high-level scripting and template metaprogramming, can enable large performance gains, while managing complexity for the developer. As the scale of available computational power continues to expand, our approach holds great potential both for accelerating progress in artificial vision, and for generating new, experimentally-testable hypotheses for the study of biological vision.
   

About the speaker

1. Nicolas Pinto is a third year PhD Student in Computational Neuroscience at MIT. He is currently a member of the DiCarlo Lab at MIT, the Sinha Lab for Vision Research at MIT, and the Cox Visual Neuroscience Group at Harvard/Rowland. His research interests lie at the intersection of Brain and Computer Sciences.
   

2. Before coming to the United States, Nicolas studied in Brazil (PUC-Rio), South Korea (CBNU), Switzerland (CERN) and received two M.S. in Computer Science (Software Engineering and Artificial Intelligence) from the University of Technology of Belfort-Montbeliard (UTBM, France) and the Ecole Nationale Supérieure d'Ingénieurs Sud Alsace (ENSISA/UHA, France).