Slow Intelligence System (SIS) 
In this talk I will introduce the concept of slow intelligence. Not all intelligent systems have fast intelligence. There are a surprisingly large number of intelligent systems, quasi-intelligent systems and semi-intelligent systems that have slow intelligence. Such slow intelligence systems are often neglected in mainstream research on intelligent systems, but they are really worthy of our attention and emulation. I will discuss the general characteristics of slow intelligence systems and then concentrate on evolutionary query processing for distributed multimedia systems as an example of artificial slow intelligence systems. …
Online ICA
Solving statistical learning problems often involves nonconvex optimization. Despite the empirical success of nonconvex statistical optimization methods, their global dynamics, especially convergence to the desirable local minima, remain less well understood in theory. In this paper, we propose a new analytic paradigm based on diffusion processes to characterize the global dynamics of nonconvex statistical optimization. As a concrete example, we study stochastic gradient descent (SGD) for the tensor decomposition formulation of independent component analysis. In particular, we cast different phases of SGD into diffusion processes, i.e., solutions to stochastic differential equations. Initialized from an unstable equilibrium, the global dynamics of SGD transit over three consecutive phases: (i) an unstable Ornstein-Uhlenbeck process slowly departing from the initialization, (ii) the solution to an ordinary differential equation, which quickly evolves towards the desirable local minimum, and (iii) a stable Ornstein-Uhlenbeck process oscillating around the desirable local minimum. Our proof techniques are based upon Stroock and Varadhan’s weak convergence of Markov chains to diffusion processes, which are of independent interest. …
DiracNet
Deep neural networks with skip-connections, such as ResNet, show excellent performance in various image classification benchmarks. It is though observed that the initial motivation behind them – training deeper networks – does not actually hold true, and the benefits come from increased capacity, rather than from depth. Motivated by this, and inspired from ResNet, we propose a simple Dirac weight parameterization, which allows us to train very deep plain networks without skip-connections, and achieve nearly the same performance. This parameterization has a minor computational cost at training time and no cost at all at inference. We’re able to achieve 95.5% accuracy on CIFAR-10 with 34-layer deep plain network, surpassing 1001-layer deep ResNet, and approaching Wide ResNet. Our parameterization also mostly eliminates the need of careful initialization in residual and non-residual networks. The code and models for our experiments are available at https://…/diracnets …
Like this:
Like Loading…
Related