Deep learning is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. Because the computer gathers knowledge from experience, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The hierarchy of concepts allows the computer to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep. This book introduces a broad range of topics in deep learning. The text offers mathematical and conceptual background, covering relevant concepts in linear algebra, probability theory and information theory, numerical computation, and machine learning. It describes deep learning techniques used by practitioners in industry, including deep feedforward networks, regularization, optimization algorithms, convolutional networks, sequence modeling, and practical methodology; and it surveys such applications as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics, and videogames. Finally, the book offers research perspectives, covering such theoretical topics as linear factor models, autoencoders, representation learning, structured probabilistic models, Monte Carlo methods, the partition function, approximate inference, and deep generative models. Deep Learning can be used by undergraduate or graduate students planning careers in either industry or research, and by software engineers who want to begin using deep learning in their products or platforms. A website offers supplementary material for both readers and instructors.

Deep Learning

A key challenge for neural modeling is to explain how a continuous stream of multimodal input from a rapidly changing environment can be processed by stereotypical recurrent circuits of integrate-and-fire neurons in real time. We propose a new computational model for real-time computing on time-varying input that provides an alternative to paradigms based on Turing machines or attractor neural networks. It does not require a task-dependent construction of neural circuits. Instead, it is based on principles of high-dimensional dynamical systems in combination with statistical learning theory and can be implemented on generic evolved or found recurrent circuitry. It is shown that the inherent transient dynamics of the high-dimensional dynamical system formed by a sufficiently large and heterogeneous neural circuit may serve as universal analog fading memory. Readout neurons can learn to extract in real time from the current state of such recurrent neural circuit information about current and past inputs that may be needed for diverse tasks. Stable internal states are not required for giving a stable output, since transient internal states can be transformed by readout neurons into stable target outputs due to the high dimensionality of the dynamical system. Our approach is based on a rigorous computational model, the liquid state machine, that, unlike Turing machines, does not require sequential transitions between well-defined discrete internal states. It is supported, as the Turing machine is, by rigorous mathematical results that predict universal computational power under idealized conditions, but for the biologically more realistic scenario of real-time processing of time-varying inputs. Our approach provides new perspectives for the interpretation of neural coding, the design of experiments and data analysis in neurophysiology, and the solution of problems in robotics and neurotechnology.

/pdf/real-time-computing-without-stable-states-a-new-framework-p44jz35cim.pdf

Real-time computing without stable states: a new framework for neural computation based on perturbations

Many current neurophysiological, psychophysical, and psychological approaches to vision rest on the idea that when we see, the brain produces an internal representation of the world. The activation of this internal representation is assumed to give rise to the experience of seeing. The problem with this kind of approach is that it leaves unexplained how the existence of such a detailed internal representation might produce visual consciousness. An alternative proposal is made here. We propose that seeing is a way of acting. It is a particular way of exploring the environment. Activity in internal representations does not generate the experience of seeing. The outside world serves as its own, external, representation. The experience of seeing occurs when the organism masters what we call the governing laws of sensorimotor contingency. The advantage of this approach is that it provides a natural and principled way of accounting for visual consciousness, and for the differences in the perceived quality of sensory experience in the different sensory modalities. Several lines of empirical evidence are brought forward in support of the theory, in particular: evidence from experiments in sensorimotor adaptation, visual \"filling in,\" visual stability despite eye movements, change blindness, sensory substitution, and color perception.

A sensorimotor account of vision and visual consciousness-Authors' Response-Acting out our sensory experience

Many current neurophysiological, psychophysical, and psychological approaches to vision rest on the idea that when we see, the brain produces an internal representation of the world. The activation of this internal representation is assumed to give rise to the experience of seeing. The problem with this kind of approach is that it leaves unexplained how the existence of such a detailed internal representation might produce visual consciousness. An alternative proposal is made here. We propose that seeing is a way of acting. It is a particular way of exploring the environment. Activity in internal representations does not generate the experience of seeing. The outside world serves as its own, external, representation. The experience of seeing occurs when the organism masters what we call the governing laws of sensorimotor contingency. The advantage of this approach is that it provides a natural and principled way of accounting for visual consciousness, and for the differences in the perceived quality of sensory experience in the different sensory modalities. Several lines of empirical evidence are brought forward in support of the theory, in particular: evidence from experiments in sensorimotor adaptation, visual “filling in,” visual stability despite eye movements, change blindness, sensory substitution, and color perception.

/pdf/a-sensorimotor-account-of-vision-and-visual-consciousness-1o8kf5szfs.pdf

A sensorimotor account of vision and visual consciousness

Neuroeconomics uses knowledge about brain mechanisms to inform economic analysis, and roots economics in biology. It opens up the "black box" of the brain, much as organizational economics adds detail to the theory of the firm. Neuroscientists use many tools— including brain imaging, behavior of patients with localized brain lesions, animal behavior, and recording single neuron activity. The key insight for economics is that the brain is composed of multiple systems which interact. Controlled systems ("executive function") interrupt automatic ones. Emotions and cognition both guide decisions. Just as prices and allocations emerge from the interaction of two processes—supply and demand— individual decisions can be modeled as the result of two (or more) processes interacting. Indeed, "dual-process" models of this sort are better rooted in neuroscientific fact, and more empirically accurate, than single-process models (such as utility-maximization). We discuss how brain evidence complicates standard assumptions about basic preference, to include homeostasis and other kinds of state-dependence. We also discuss applications to intertemporal choice, risk and decision making, and game theory. Intertemporal choice appears to be domain-specific and heavily influenced by emotion. The simplified s-d of quasi-hyperbolic discounting is supported by activation in distinct regions of limbic and cortical systems. In risky decision, imaging data tentatively support the idea that gains and losses are coded separately, and that ambiguity is distinct from risk, because it activates fear and discomfort regions. (Ironically, lesion patients who do not receive fear signals in prefrontal cortex are "rationally" neutral toward ambiguity.) Game theory studies show the effect of brain regions implicated in "theory of mind", correlates of strategic skill, and effects of hormones and other biological variables. Finally, economics can contribute to neuroscience because simple rational-choice models are useful for understanding highly-evolved behavior like motor actions that earn rewards, and Bayesian integration of sensorimotor information.

/pdf/neuroeconomics-how-neuroscience-can-inform-economics-39l750mxro.pdf

Neuroeconomics: How Neuroscience Can Inform Economics

An unresolved issue in cortical development concerns the relative contributions of intrinsic and extrinsic factors to the functional specification of different cortical areas. Ferrets in which retinal projections are redirected neonatally to the auditory thalamus have visually responsive cells in auditory thalamus and cortex, form a retinotopic map in auditory cortex and have visual receptive field properties in auditory cortex that are typical of cells in visual cortex. Here we report that this cross-modal projection and its representation in auditory cortex can mediate visual behaviour. When light stimuli are presented in the portion of the visual field that is 'seen' only by this projection, 'rewired' ferrets respond as though they perceive the stimuli to be visual rather than auditory. Thus the perceptual modality of a neocortical region is instructed to a significant extent by its extrinsic inputs. In addition, gratings of different spatial frequencies can be discriminated by the rewired pathway, although the grating acuity is lower than that of the normal visual pathway.

Laurie von Melchner

Papers

Visual behaviour mediated by retinal projections directed to the auditory pathway