Robert Coop

In this paper we present the fixed expansion layer (FEL) feedforward neural network designed for balancing plasticity and stability in the presence of non-stationary inputs. Catastrophic interference (or catastrophic forgetting) refers to the drastic loss of previously learned information when a neural network is trained on new or different information. The goal of the FEL network is to reduce the effect of catastrophic interference by augmenting a multilayer perceptron with a layer of sparse… Show more

In this paper we present the fixed expansion layer (FEL) feedforward neural network designed for balancing plasticity and stability in the presence of non-stationary inputs. Catastrophic interference (or catastrophic forgetting) refers to the drastic loss of previously learned information when a neural network is trained on new or different information. The goal of the FEL network is to reduce the effect of catastrophic interference by augmenting a multilayer perceptron with a layer of sparse neurons with binary activations. We compare the FEL network's performance to that of other algorithms designed to combat the effects of catastrophic interference and demonstrate that the FEL network is able to retain information for significantly longer periods of time with substantially lower computational requirements. Show less

We present the broad outlines of a roadmap toward humanlevel artificial general intelligence (henceforth, AGI). We begin by discussing AGI in general, adopting a pragmatic goal for its attainment and a necessary foundation of characteristics and requirements. An initial capability landscape will be presented, drawing on major themes from developmental psychology and illuminated by mathematical, physiological and information processing perspectives. The challenge of identifying appropriate tasks… Show more

We present the broad outlines of a roadmap toward humanlevel artificial general intelligence (henceforth, AGI). We begin by discussing AGI in general, adopting a pragmatic goal for its attainment and a necessary foundation of characteristics and requirements. An initial capability landscape will be presented, drawing on major themes from developmental psychology and illuminated by mathematical, physiological and information processing perspectives. The challenge of identifying appropriate tasks and environments for measuring AGI will be addressed, and seven scenarios will be presented as milestones suggesting a roadmap across the AGI landscape along with directions for future research and collaboration. Show less

This work introduces a method for predicting tumor growth by:
1) Identifying functional masks for the tumor with high predictive power
2) Learning parameters that predict growth based on these masks
Given simulated tumor growth trajectories (presented as cancer cell counts for voxels within a 128x128 grid), we first examine a large number of trajectories in order to determine, for any given central voxel, which neighboring voxels have the
highest predictive power in… Show more

This work introduces a method for predicting tumor growth by:
1) Identifying functional masks for the tumor with high predictive power
2) Learning parameters that predict growth based on these masks
Given simulated tumor growth trajectories (presented as cancer cell counts for voxels within a 128x128 grid), we first examine a large number of trajectories in order to determine, for any given central voxel, which neighboring voxels have the
highest predictive power in determinining future states of the central voxel (we refer to these sets of predictive voxels as 'functional masks'). A novel functional analysis procedure is developed which uses an evolutionary algorithm combined with general systems theory principles in order to search for highly predictive functional masks. Probabilistic parameter estimation is then performed over the voxels in this
functional mask in order to determine the probability distribution of the next state of the central voxel conditioned on the current and previous states of its neighbors.

Using this approach we are able to make accurate predictions of tumor growth using general methods that do not rely on biological or other domain­ specific knowledge about the detailed mechanisms underlying the growth. Show less

The topic of deep learning systems has received significant attention during the past few years, particularly as a biologically-inspired approach to processing high-
dimensional signals. The latter often involve spatiotemporal information that may span large scales, rendering its representation in the general case highly challenging. Deep learning networks attempt to overcome this challenge by means of a hierarchical architecture that is comprised of common
circuits with similar (and… Show more

The topic of deep learning systems has received significant attention during the past few years, particularly as a biologically-inspired approach to processing high-
dimensional signals. The latter often involve spatiotemporal information that may span large scales, rendering its representation in the general case highly challenging. Deep learning networks attempt to overcome this challenge by means of a hierarchical architecture that is comprised of common
circuits with similar (and often cortically influenced) functionality. The goal of such systems is to represent sensory
observations in a manner that will later facilitate robust pattern classification, mimicking a key attribute of the mammal
brain. This stands in contrast with the mainstream approach of pre-processing the data so as to reduce its dimensionality
- a paradigm that often results in sub-optimal performance.
This paper presents a Deep SpatioTemporal Inference Network (DeSTIN)- ascalable deeplearning architecture that relies on a combination of unsupervised learning and Bayesian inference. Dynamic pattern learning forms an inherent way of capturing complex spatiotemporal dependencies. Simulation results demonstrate the core capabilities of the proposed framework, particularly in the context of high-dimensional signal classification. Show less