Deep neural networks (DNNs) have demonstrated state-of-the-art results on many pattern recognition tasks, especially vision classification problems. Understanding the inner workings of such computational brains is both fascinating basic science that is interesting in its own right---similar to why we study the human brain---and will enable researchers to further improve DNNs. One path to understanding how a neural network functions internally is to study what each of its neurons has learned to detect. One such method is called activation maximization, which synthesizes an input (e.g. an image) that highly activates a neuron. Here we dramatically improve the qualitative state of the art of activation maximization by harnessing a powerful, learned prior: a deep generator network. The algorithm (1) generates qualitatively state-of-the-art synthetic images that look almost real, (2) reveals the features learned by each neuron in an interpretable way, (3) generalizes well to new datasets and somewhat well to different network architectures without requiring the prior to be relearned, and (4) can be considered as a high-quality generative method (in this case, by generating novel, creative, interesting, recognizable images).

Deep neural networks (DNNs) have recently been achieving state-of-the-art performance on a variety of pattern-recognition tasks, most notably visual classification problems. Given that DNNs are now able to classify objects in images with near-human-level performance, questions naturally arise as to what differences remain between computer and human vision. A recent study revealed that changing an image (e.g. of a lion) in a way imperceptible to humans can cause a DNN to label the image as something else entirely (e.g. mislabeling a lion a library). Here we show a related result: it is easy to produce images that are completely unrecognizable to humans, but that state-of-the-art DNNs believe to be recognizable objects with 99.99% confidence (e.g. labeling with certainty that white noise static is a lion). Specifically, we take convolutional neural networks trained to perform well on either the ImageNet or MNIST datasets and then find images with evolutionary algorithms or gradient ascent that DNNs label with high confidence as belonging to each dataset class. It is possible to produce images totally unrecognizable to human eyes that DNNs believe with near certainty are familiar objects, which we call "fooling images" (more generally, fooling examples). Our results shed light on interesting differences between human vision and current DNNs, and raise questions about the generality of DNN computer vision.

Many deep neural networks trained on natural images exhibit a curious phenomenon in common: on the first layer they learn features similar to Gabor filters and color blobs. Such first-layer features appear not to be specific to a particular dataset or task, but general in that they are applicable to many datasets and tasks. Features must eventually transition from general to specific by the last layer of the network, but this transition has not been studied extensively. In this paper we experimentally quantify the generality versus specificity of neurons in each layer of a deep convolutional neural network and report a few surprising results. Transferability is negatively affected by two distinct issues: (1) the specialization of higher layer neurons to their original task at the expense of performance on the target task, which was expected, and (2) optimization difficulties related to splitting networks between co-adapted neurons, which was not expected. In an example network trained on ImageNet, we demonstrate that either of these two issues may dominate, depending on whether features are transferred from the bottom, middle, or top of the network. We also document that the transferability of features decreases as the distance between the base task and target task increases, but that transferring features even from distant tasks can be better than using random features. A final surprising result is that initializing a network with transferred features from almost any number of layers can produce a boost to generalization that lingers even after fine-tuning to the target dataset.

The Association for the Advancement of Artificial Intelligence was pleased to present the 2013 Fall Symposium Series, held Friday through Sunday, November 15–17, at the Westin Arlington Gateway in Arlington, Virginia near Washington DC USA. The titles of the five symposia were as follows: Discovery Informatics: AI Takes a Science-Centered View on Big Data (FS-13-01); How Should Intelligence be Abstracted in AI Research: MDPs, Symbolic Representations, Artificial Neural Networks, or --? The highlights of each symposium are presented in this report.

Rinke Hoekstra (VU University from transferring and adapting semantic web Amsterdam) presented linked open data tools technologies to the big data quest. Finally, in the Social to discover connections within established scientific Networks and Social Contagion symposium, a data sets. Louiqa Rashid (University of Maryland) community of researchers explored topics such as social presented work on similarity metrics linking together contagion, game theory, network modeling, network-based drugs, genes, and diseases. Kyle Ambert (Intel) presented inference, human data elicitation, and Finna, a text-mining system to identify passages web analytics. Highlights of the symposia are contained of interest containing descriptions of neuronal in this report.

Subfields of artificial intelligence often diversify from a core idea. For example, deep learning networks, models in computational neuroscience, and neuroevolution all take inspiration from biological neural networks as a potential pathway to AI. Most researchers choose to pursue the subfield (and by extension, abstraction) they see as most promising for leading to AI, which naturally results in significant debate and disagreement among researchers as to what abstraction is best. A better understanding and less polarized debate may result from a clear presentation and discussion of abstractions by their most knowledgeable proponents. These insights motivated bringing together researchers from fields that abstract AI at different levels or in different ways to disperse knowledge, and to critically examining the value and promise of different abstractions. Thus this AAAI symposium, How Intelligence Should be Abstracted in AI, consisted of a diverse and multidisciplinary group of AI researchers interested in discussing and comparing different abstractions of both intelligence and processes that might create it.

List of organizers of the AAAI Fall Symposium, How SHould Intelligence Be Abstracted in AI Research: MDPs, Symbolic Representations, Artificial Neural Networks, or _____?