TY - JOUR
TI - Analogues of mental simulation and imagination in deep learning
AU - Hamrick, Jessica B
T2 - Current Opinion in Behavioral Sciences
T3 - SI: 29: Artificial Intelligence (2019)
AB - Mental simulation—the capacity to imagine what will or what could be—is a salient feature of human cognition, playing a key role in a wide range of cognitive abilities. In artificial intelligence, the last few years have seen the development of methods which are analogous to mental models and mental simulation. This paper outlines recent methods in deep learning for constructing such models from data and learning to use them via reinforcement learning, and compares such approaches to human mental simulation. Model-based methods in deep learning can serve as powerful tools for building and scaling cognitive models. However, a number of challenges remain in matching the capacity of human mental simulation for efficiency, compositionality, generalization, and creativity.
DA - 2019/10/01/
PY - 2019
DO - 10.1016/j.cobeha.2018.12.011
DP - ScienceDirect
VL - 29
SP - 8
EP - 16
J2 - Current Opinion in Behavioral Sciences
SN - 2352-1546
UR - http://www.sciencedirect.com/science/article/pii/S2352154618301670
Y2 - 2019/10/10/19:15:54
ER -
TY - BOOK
TI - Model-Based Machine Learning
AU - Winn, John Michael
AB - This book is unusual for a machine learning text book in that the authors do not review dozens of different algorithms. Instead they introduce all of the key ideas through a series of case studies involving real-world applications. Case studies play a central role because it is only in the context of applications that it makes sense to discuss modelling assumptions. Each chapter therefore introduces one case study which is drawn from a real-world application that has been solved using a model-based approach.
DA - 2019/06//
PY - 2019
DP - Google Books
SP - 400
LA - en
PB - Taylor & Francis Incorporated
SN - 978-1-4987-5681-5
KW - Bayesian inference
KW - Classical ML
KW - Implementation
ER -
TY - JOUR
TI - Relational inductive biases, deep learning, and graph networks
AU - Battaglia, Peter W.
AU - Hamrick, Jessica B.
AU - Bapst, Victor
AU - Sanchez-Gonzalez, Alvaro
AU - Zambaldi, Vinicius
AU - Malinowski, Mateusz
AU - Tacchetti, Andrea
AU - Raposo, David
AU - Santoro, Adam
AU - Faulkner, Ryan
AU - Gulcehre, Caglar
AU - Song, Francis
AU - Ballard, Andrew
AU - Gilmer, Justin
AU - Dahl, George
AU - Vaswani, Ashish
AU - Allen, Kelsey
AU - Nash, Charles
AU - Langston, Victoria
AU - Dyer, Chris
AU - Heess, Nicolas
AU - Wierstra, Daan
AU - Kohli, Pushmeet
AU - Botvinick, Matt
AU - Vinyals, Oriol
AU - Li, Yujia
AU - Pascanu, Razvan
T2 - arXiv:1806.01261 [cs, stat]
AB - Artificial intelligence (AI) has undergone a renaissance recently, making major progress in key domains such as vision, language, control, and decision-making. This has been due, in part, to cheap data and cheap compute resources, which have fit the natural strengths of deep learning. However, many defining characteristics of human intelligence, which developed under much different pressures, remain out of reach for current approaches. In particular, generalizing beyond one's experiences--a hallmark of human intelligence from infancy--remains a formidable challenge for modern AI. The following is part position paper, part review, and part unification. We argue that combinatorial generalization must be a top priority for AI to achieve human-like abilities, and that structured representations and computations are key to realizing this objective. Just as biology uses nature and nurture cooperatively, we reject the false choice between "hand-engineering" and "end-to-end" learning, and instead advocate for an approach which benefits from their complementary strengths. We explore how using relational inductive biases within deep learning architectures can facilitate learning about entities, relations, and rules for composing them. We present a new building block for the AI toolkit with a strong relational inductive bias--the graph network--which generalizes and extends various approaches for neural networks that operate on graphs, and provides a straightforward interface for manipulating structured knowledge and producing structured behaviors. We discuss how graph networks can support relational reasoning and combinatorial generalization, laying the foundation for more sophisticated, interpretable, and flexible patterns of reasoning. As a companion to this paper, we have released an open-source software library for building graph networks, with demonstrations of how to use them in practice.
DA - 2018/06/04/
PY - 2018
DP - arXiv.org
UR - http://arxiv.org/abs/1806.01261
Y2 - 2019/10/10/19:16:22
ER -
TY - JOUR
TI - Adversarial Patch
AU - Brown, Tom B.
AU - Mané, Dandelion
AU - Roy, Aurko
AU - Abadi, Martín
AU - Gilmer, Justin
T2 - arXiv:1712.09665 [cs]
AB - We present a method to create universal, robust, targeted adversarial image patches in the real world. The patches are universal because they can be used to attack any scene, robust because they work under a wide variety of transformations, and targeted because they can cause a classifier to output any target class. These adversarial patches can be printed, added to any scene, photographed, and presented to image classifiers; even when the patches are small, they cause the classifiers to ignore the other items in the scene and report a chosen target class. To reproduce the results from the paper, our code is available at https://github.com/tensorflow/cleverhans/tree/master/examples/adversarial_patch
DA - 2018/05/16/
PY - 2018
DP - arXiv.org
UR - http://arxiv.org/abs/1712.09665
Y2 - 2019/11/23/14:10:12
KW - Adversarial attacks
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Robust Physical-World Attacks on Deep Learning Models
AU - Eykholt, Kevin
AU - Evtimov, Ivan
AU - Fernandes, Earlence
AU - Li, Bo
AU - Rahmati, Amir
AU - Xiao, Chaowei
AU - Prakash, Atul
AU - Kohno, Tadayoshi
AU - Song, Dawn
T2 - arXiv:1707.08945 [cs]
AB - Recent studies show that the state-of-the-art deep neural networks (DNNs) are vulnerable to adversarial examples, resulting from small-magnitude perturbations added to the input. Given that that emerging physical systems are using DNNs in safety-critical situations, adversarial examples could mislead these systems and cause dangerous situations.Therefore, understanding adversarial examples in the physical world is an important step towards developing resilient learning algorithms. We propose a general attack algorithm,Robust Physical Perturbations (RP2), to generate robust visual adversarial perturbations under different physical conditions. Using the real-world case of road sign classification, we show that adversarial examples generated using RP2 achieve high targeted misclassification rates against standard-architecture road sign classifiers in the physical world under various environmental conditions, including viewpoints. Due to the current lack of a standardized testing method, we propose a two-stage evaluation methodology for robust physical adversarial examples consisting of lab and field tests. Using this methodology, we evaluate the efficacy of physical adversarial manipulations on real objects. Witha perturbation in the form of only black and white stickers,we attack a real stop sign, causing targeted misclassification in 100% of the images obtained in lab settings, and in 84.8%of the captured video frames obtained on a moving vehicle(field test) for the target classifier.
DA - 2018/04/10/
PY - 2018
DP - arXiv.org
UR - http://arxiv.org/abs/1707.08945
Y2 - 2019/11/23/14:08:00
KW - Adversarial attacks
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Automatic differentiation in machine learning: a survey
AU - Baydin, Atilim Gunes
AU - Pearlmutter, Barak A.
AU - Radul, Alexey Andreyevich
AU - Siskind, Jeffrey Mark
T2 - arXiv:1502.05767 [cs, stat]
AB - Derivatives, mostly in the form of gradients and Hessians, are ubiquitous in machine learning. Automatic differentiation (AD), also called algorithmic differentiation or simply "autodiff", is a family of techniques similar to but more general than backpropagation for efficiently and accurately evaluating derivatives of numeric functions expressed as computer programs. AD is a small but established field with applications in areas including computational fluid dynamics, atmospheric sciences, and engineering design optimization. Until very recently, the fields of machine learning and AD have largely been unaware of each other and, in some cases, have independently discovered each other's results. Despite its relevance, general-purpose AD has been missing from the machine learning toolbox, a situation slowly changing with its ongoing adoption under the names "dynamic computational graphs" and "differentiable programming". We survey the intersection of AD and machine learning, cover applications where AD has direct relevance, and address the main implementation techniques. By precisely defining the main differentiation techniques and their interrelationships, we aim to bring clarity to the usage of the terms "autodiff", "automatic differentiation", and "symbolic differentiation" as these are encountered more and more in machine learning settings.
DA - 2018/02/05/
PY - 2018
DP - arXiv.org
ST - Automatic differentiation in machine learning
UR - http://arxiv.org/abs/1502.05767
Y2 - 2019/11/22/22:28:45
KW - Automatic differentiation
KW - Classical ML
KW - Differentiation
KW - Machine learning
ER -
TY - JOUR
TI - Understanding deep learning requires rethinking generalization
AU - Zhang, Chiyuan
AU - Bengio, Samy
AU - Hardt, Moritz
AU - Recht, Benjamin
AU - Vinyals, Oriol
T2 - arXiv:1611.03530 [cs]
AB - Despite their massive size, successful deep artificial neural networks can exhibit a remarkably small difference between training and test performance. Conventional wisdom attributes small generalization error either to properties of the model family, or to the regularization techniques used during training. Through extensive systematic experiments, we show how these traditional approaches fail to explain why large neural networks generalize well in practice. Specifically, our experiments establish that state-of-the-art convolutional networks for image classification trained with stochastic gradient methods easily fit a random labeling of the training data. This phenomenon is qualitatively unaffected by explicit regularization, and occurs even if we replace the true images by completely unstructured random noise. We corroborate these experimental findings with a theoretical construction showing that simple depth two neural networks already have perfect finite sample expressivity as soon as the number of parameters exceeds the number of data points as it usually does in practice. We interpret our experimental findings by comparison with traditional models.
DA - 2017/02/26/
PY - 2017
DP - arXiv.org
UR - http://arxiv.org/abs/1611.03530
Y2 - 2019/11/22/20:11:42
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Adversarial examples in the physical world
AU - Kurakin, Alexey
AU - Goodfellow, Ian
AU - Bengio, Samy
T2 - arXiv:1607.02533 [cs, stat]
AB - Most existing machine learning classifiers are highly vulnerable to adversarial examples. An adversarial example is a sample of input data which has been modified very slightly in a way that is intended to cause a machine learning classifier to misclassify it. In many cases, these modifications can be so subtle that a human observer does not even notice the modification at all, yet the classifier still makes a mistake. Adversarial examples pose security concerns because they could be used to perform an attack on machine learning systems, even if the adversary has no access to the underlying model. Up to now, all previous work have assumed a threat model in which the adversary can feed data directly into the machine learning classifier. This is not always the case for systems operating in the physical world, for example those which are using signals from cameras and other sensors as an input. This paper shows that even in such physical world scenarios, machine learning systems are vulnerable to adversarial examples. We demonstrate this by feeding adversarial images obtained from cell-phone camera to an ImageNet Inception classifier and measuring the classification accuracy of the system. We find that a large fraction of adversarial examples are classified incorrectly even when perceived through the camera.
DA - 2017/02/10/
PY - 2017
DP - arXiv.org
UR - http://arxiv.org/abs/1607.02533
Y2 - 2019/11/23/14:08:43
KW - Adversarial attacks
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Attention and Augmented Recurrent Neural Networks
AU - Olah, Chris
AU - Carter, Shan
T2 - Distill
AB - A visual overview of neural attention, and the powerful extensions of neural networks being built on top of it.
DA - 2016/09/08/
PY - 2016
DO - 10/gf33sg
DP - distill.pub
VL - 1
IS - 9
SP - e1
J2 - Distill
LA - en
SN - 2476-0757
UR - http://distill.pub/2016/augmented-rnns
Y2 - 2019/11/22/20:09:48
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Explaining and Harnessing Adversarial Examples
AU - Goodfellow, Ian J.
AU - Shlens, Jonathon
AU - Szegedy, Christian
T2 - arXiv:1412.6572 [cs, stat]
AB - Several machine learning models, including neural networks, consistently misclassify adversarial examples---inputs formed by applying small but intentionally worst-case perturbations to examples from the dataset, such that the perturbed input results in the model outputting an incorrect answer with high confidence. Early attempts at explaining this phenomenon focused on nonlinearity and overfitting. We argue instead that the primary cause of neural networks' vulnerability to adversarial perturbation is their linear nature. This explanation is supported by new quantitative results while giving the first explanation of the most intriguing fact about them: their generalization across architectures and training sets. Moreover, this view yields a simple and fast method of generating adversarial examples. Using this approach to provide examples for adversarial training, we reduce the test set error of a maxout network on the MNIST dataset.
DA - 2015/03/20/
PY - 2015
DP - arXiv.org
UR - http://arxiv.org/abs/1412.6572
Y2 - 2019/11/23/14:10:23
KW - Adversarial attacks
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Why does Deep Learning work? - A perspective from Group Theory
AU - Paul, Arnab
AU - Venkatasubramanian, Suresh
T2 - arXiv:1412.6621 [cs, stat]
AB - Why does Deep Learning work? What representations does it capture? How do higher-order representations emerge? We study these questions from the perspective of group theory, thereby opening a new approach towards a theory of Deep learning. One factor behind the recent resurgence of the subject is a key algorithmic step called pre-training: first search for a good generative model for the input samples, and repeat the process one layer at a time. We show deeper implications of this simple principle, by establishing a connection with the interplay of orbits and stabilizers of group actions. Although the neural networks themselves may not form groups, we show the existence of {\em shadow} groups whose elements serve as close approximations. Over the shadow groups, the pre-training step, originally introduced as a mechanism to better initialize a network, becomes equivalent to a search for features with minimal orbits. Intuitively, these features are in a way the {\em simplest}. Which explains why a deep learning network learns simple features first. Next, we show how the same principle, when repeated in the deeper layers, can capture higher order representations, and why representation complexity increases as the layers get deeper.
DA - 2015/02/28/
PY - 2015
DP - arXiv.org
ST - Why does Deep Learning work?
UR - http://arxiv.org/abs/1412.6621
Y2 - 2019/11/22/17:38:08
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Probabilistic machine learning and artificial intelligence
AU - Ghahramani, Zoubin
T2 - Nature
DA - 2015/05//
PY - 2015
DO - 10/gdxwhq
DP - Crossref
VL - 521
IS - 7553
SP - 452
EP - 459
LA - en
SN - 0028-0836, 1476-4687
UR - http://www.nature.com/articles/nature14541
Y2 - 2019/11/28/12:16:49
KW - Bayesian inference
KW - Classical ML
KW - Machine learning
KW - Probabilistic programming
ER -
TY - JOUR
TI - Neural Turing Machines
AU - Graves, Alex
AU - Wayne, Greg
AU - Danihelka, Ivo
T2 - arXiv:1410.5401 [cs]
AB - We extend the capabilities of neural networks by coupling them to external memory resources, which they can interact with by attentional processes. The combined system is analogous to a Turing Machine or Von Neumann architecture but is differentiable end-to-end, allowing it to be efficiently trained with gradient descent. Preliminary results demonstrate that Neural Turing Machines can infer simple algorithms such as copying, sorting, and associative recall from input and output examples.
DA - 2014/12/10/
PY - 2014
DP - arXiv.org
UR - http://arxiv.org/abs/1410.5401
Y2 - 2019/11/21/21:09:35
KW - Abstract machines
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - Generative Adversarial Networks
AU - Goodfellow, Ian J.
AU - Pouget-Abadie, Jean
AU - Mirza, Mehdi
AU - Xu, Bing
AU - Warde-Farley, David
AU - Ozair, Sherjil
AU - Courville, Aaron
AU - Bengio, Yoshua
T2 - arXiv:1406.2661 [cs, stat]
AB - We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a discriminative model D that estimates the probability that a sample came from the training data rather than G. The training procedure for G is to maximize the probability of D making a mistake. This framework corresponds to a minimax two-player game. In the space of arbitrary functions G and D, a unique solution exists, with G recovering the training data distribution and D equal to 1/2 everywhere. In the case where G and D are defined by multilayer perceptrons, the entire system can be trained with backpropagation. There is no need for any Markov chains or unrolled approximate inference networks during either training or generation of samples. Experiments demonstrate the potential of the framework through qualitative and quantitative evaluation of the generated samples.
DA - 2014/06/10/
PY - 2014
DP - arXiv.org
UR - http://arxiv.org/abs/1406.2661
Y2 - 2019/11/28/11:44:28
KW - Adversarial attacks
KW - Classical ML
KW - Implementation
KW - Machine learning
ER -
TY - CHAP
TI - Tomaso A. Poggio autobiography
AU - Poggio, Tomaso
DA - 2013///
PY - 2013
SP - 54
UR - http://poggio-lab.mit.edu/sites/default/files/cv/tomasopoggio.pdf
KW - Classical ML
KW - Compendium
KW - Machine learning
ER -
TY - CONF
TI - Machine Learning Biochemical Networks from Temporal Logic Properties
AU - Fages, François
AU - Calzone, Laurence
AU - Chabrier-Rivier, Nathalie
AU - Soliman, Sylvain
A2 - Priami, Corrado
A2 - Plotkin, Gordon
T3 - Lecture Notes in Computer Science
AB - One central issue in systems biology is the definition of formal languages for describing complex biochemical systems and their behavior at different levels. The biochemical abstract machine BIOCHAM is based on two formal languages, one rule-based language used for modeling biochemical networks, at three abstraction levels corresponding to three semantics: boolean, concentration and population; and one temporal logic language used for formalizing the biological properties of the system. In this paper, we show how the temporal logic language can be turned into a specification language. We describe two algorithms for inferring reaction rules and kinetic parameter values from a temporal specification formalizing the biological data. Then, with an example of the cell cycle control, we illustrate how these machine learning techniques may be useful to the modeler.
C1 - Berlin, Heidelberg
C3 - Transactions on Computational Systems Biology VI
DA - 2006///
PY - 2006
DO - 10/dd8
DP - Springer Link
SP - 68
EP - 94
LA - en
PB - Springer
SN - 978-3-540-46236-1
KW - Abstract machines
KW - Biology
KW - Classical ML
KW - Machine learning
KW - Symbolic logic
KW - Systems biology
ER -
TY - CHAP
TI - Graphical Models: Overview
AU - Wermuth, N.
AU - Cox, D. R.
T2 - International Encyclopedia of the Social & Behavioral Sciences
A2 - Smelser, Neil J.
A2 - Baltes, Paul B.
AB - Graphical Markov models provide a method of representing possibly complicated multivariate dependencies in such a way that the general qualitative features can be understood, that statistical independencies are highlighted, and that some properties can be derived directly. Variables are represented by the nodes of a graph. Pairs of nodes may be joined by an edge. Edges are directed if one variable is a response to the other variable considered as explanatory, but are undirected if the variables are on an equal footing. Absence of an edge typically implies statistical independence, conditional, or marginal depending on the kind of graph. The need for a number of types of graph arises because it is helpful to represent a number of different kinds of dependence structures. Of special importance are chain graphs in which variables are arranged in a sequence or chain of blocks, the variables in any one block being on an equal footing, some being possibly joint responses to variables in the past and some being jointly explanatory to variables in the future of the block considered. Some main properties of such systems are outlined, and recent research results are sketched. Suggestions for further reading are given. As an illustrative example, some analysis of data on the treatment of chronic pain is presented.
CY - Oxford
DA - 2001/01/01/
PY - 2001
DP - ScienceDirect
SP - 6379
EP - 6386
LA - en
PB - Pergamon
SN - 978-0-08-043076-8
ST - Graphical Models
UR - http://www.sciencedirect.com/science/article/pii/B008043076700440X
Y2 - 2019/11/22/19:12:23
KW - Bayesianism
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - A Tutorial on Learning With Bayesian Networks
AU - Heckerman, David
AB - A Bayesian network is a graphical model that encodes probabilistic relationships among variables of interest. When used in conjunction with statistical techniques, the graphical model has several advantages for data analysis. One, because the model encodes dependencies among all variables, it readily handles situations where some data entries are missing. Two, a Bayesian network can …
DA - 1995/03/01/
PY - 1995
DP - www.microsoft.com
LA - en-US
UR - https://www.microsoft.com/en-us/research/publication/a-tutorial-on-learning-with-bayesian-networks/
Y2 - 2019/11/22/19:09:15
KW - Bayesianism
KW - Classical ML
KW - Machine learning
ER -
TY - JOUR
TI - On the Computational Power of Neural Nets
AU - Siegelmann, H. T.
AU - Sontag, E. D.
T2 - Journal of Computer and System Sciences
AB - This paper deals with finite size networks which consist of interconnections of synchronously evolving processors. Each processor updates its state by applying a "sigmoidal" function to a linear combination of the previous states of all units. We prove that one may simulate all Turing machines by such nets. In particular, one can simulate any multi-stack Turing machine in real time, and there is a net made up of 886 processors which computes a universal partial-recursive function. Products (high order nets) are not required, contrary to what had been stated in the literature. Non-deterministic Turing machines can be simulated by non-deterministic rational nets, also in real time. The simulation result has many consequences regarding the decidability, or more generally the complexity, of questions about recursive nets.
DA - 1995/02/01/
PY - 1995
DO - 10/dvwtc3
DP - ScienceDirect
VL - 50
IS - 1
SP - 132
EP - 150
J2 - Journal of Computer and System Sciences
LA - en
SN - 0022-0000
UR - http://www.sciencedirect.com/science/article/pii/S0022000085710136
Y2 - 2019/11/28/17:50:06
KW - Classical ML
KW - Machine learning
ER -
TY - SLIDE
TI - Mathematics of AlphaGo
A2 - Murfet, Daniel
KW - Classical ML
KW - Machine learning
ER -
TY - SLIDE
TI - Algebra and Artiﬁcial Intelligence
A2 - Murfet, Daniel
LA - en
KW - Algebra
KW - Classical ML
KW - Machine learning
KW - Sketchy
ER -