Research Meeting: Module 3

Nov 12, 2019 - 4:00 pm
Venue:  MIT 46-5165 Address:  43 Vassar St, MIT bldg 46, Cambridge MA 02139 Speaker/s:  Katharina Dobs(Kanwisher Lab)

Using task-optimized neural networks to understand why brains have specialized processing for faces

Previous research has identified multiple functionally specialized regions of the human visual cortex and has started to characterize the precise function of these regions. But why do brains have functional specialization in the first place, and why do we have the particular specializations we do (e.g., for faces or scenes, but apparently not for food or cars)? Here, we address these questions using the well-studied case of face selectivity. Specifically, we used deep convolutional neural networks (CNNs) to test whether face-specific regions are segregated from object cortex in the primate visual system because the optimal feature spaces for face and object perception differ from each other.
    We trained two separate CNNs with the AlexNet architecture to categorize either faces or objects. The face-trained CNN performed worse on object categorization than the object-trained CNN and vice versa, demonstrating that the learned features differ for the two tasks. To determine whether a CNN could learn a common feature space, we trained CNNs on both tasks with a branched architecture, varying the number of layers that were shared across tasks (Kell et al., 2018). The fully-shared network performed worse than the separate CNNs suggesting a cost for sharing both tasks in one system. However, like in the primate visual system, early layers could be shared without impairing performance.

    Do these results generalize to architectures with larger capacity? We trained three networks with a VGG16 architecture: one on faces, one on objects, and one on both. In contrast to the AlexNet results, the dual-task CNN performed as well as the separate networks. To test whether this network discovered covert task segregation, we performed lesion experiments and found that lesioning face-specific features selectively impaired face performance. This result suggests that functional specialization for faces emerges spontaneously in networks optimized for face and object tasks.

Organizer:  Hector Penagos Frederico Azevedo Organizer Email:  cbmm-contact@mit.edu

Brains, Minds + Machines Seminar Series: Feedforward and feedback processes in visual recognition

Nov 5, 2019 - 4:00 pm
Photo of Thomas Serre
Venue:  Singleton Auditorium Address:  43 Vassar Street, Cambridge MA 02139 Speaker/s:  Thomas Serre, Cognitive, Linguistic & Psychological Sciences Department, Carney Institute for Brain Science, Brown University

Title: Feedforward and feedback processes in visual recognition

Abstract: Progress in deep learning has spawned great successes in many engineering applications. As a prime example, convolutional neural networks, a type of feedforward neural networks, are now approaching – and sometimes even surpassing – human accuracy on a variety of visual recognition tasks. In this talk, however, I will show that these neural networks and their recent extensions exhibit a limited ability to solve seemingly simple visual reasoning problems involving incremental grouping, similarity and spatial relation judgments. Our group has developed a recurrent network model of classical and extra-classical receptive fields that is constrained by the anatomy and physiology of the visual cortex. The model was shown to account for diverse visual illusions providing computational evidence for a novel canonical circuit that is shared across visual modalities. I will show that this computational neuroscience model can be turned into a modern end-to-end trainable deep recurrent network architecture which addresses some of the shortcomings exhibited by state-of-the-art feedforward networks for solving complex visual reasoning tasks. This suggests that neuroscience may contribute powerful new ideas and approaches to computer science and artificial intelligence.​

Organizer:  Hector Penagos Organizer Email:  cbmm-contact@mit.edu

CBMM Special Seminar: Beyond Empirical Risk Minimization: the lessons of deep learning

Oct 28, 2019 - 4:00 pm
Photo of Mikhail Belkin
Venue:  Singleton Auditorium Address:  43 Vassar Street, Cambridge MA 02139 Speaker/s:  Mikhail Belkin, Professor, The Ohio State University - Department of Computer Science and Engineering, Department of Statistics, Center for Cognitive Science

Title: Beyond Empirical Risk Minimization: the lessons of deep learning

Abstract: "A model with zero training error is  overfit to the training data and  will typically generalize poorly"  goes statistical textbook wisdom.  Yet, in modern practice, over-parametrized deep networks with   near  perfect  fit on  training data still show excellent test performance.  This apparent  contradiction points to troubling cracks in the conceptual foundations of machine learning. While classical analyses of Empirical Risk Minimization rely on balancing the  complexity of  predictors with  training error, modern models are best described by interpolation. In that paradigm  a predictor is chosen by minimizing (explicitly or implicitly) a norm corresponding to a certain inductive bias over a space of functions that  fit the training data exactly. I will discuss the nature of the challenge to our understanding of machine learning and point the way forward to first analyses that account for the empirically observed phenomena.  Furthermore, I will show how  classical and modern models can  be unified within a single  "double descent" risk curve,  which subsumes the classical U-shaped bias-variance trade-off.

Finally, as an example of a particularly interesting inductive bias, I will show evidence that deep  over-parametrized autoencoders networks, trained with SGD, implement a form of associative memory with training examples as attractor states.

Organizer:  Jean Lawrence Organizer Email:  cbmm-contact@mit.edu

Pages

Subscribe to Front page feed