A network underlying human higher-order motor control: Insights from machine learning-based lesion-behaviour mapping in apraxia of pantomime

Abstract According to the theory of contextual integrity (CI), privacy norms prescribe information flows with reference to five parameters — sender, recipient, subject, information type, and transmission principle. Because privacy is grasped contextually (e.g., health, education, civic life, etc.), the values of these parameters range over contextually meaningful ontologies — of information types (or topics) and actors (subjects, senders, and recipients), in contextually defined capacities. As an alternative to predominant approaches to privacy, which were ineffective against novel information practices enabled by IT, CI was able both to pinpoint sources of disruption and provide grounds for either accepting or rejecting them. Mounting challenges from a burgeoning array of networked, sensor-enabled devices (IoT) and data-ravenous machine learning systems, similar in form though magnified in scope, call for renewed attention to theory. This Article introduces the metaphor of a data (food) chain to capture the nature of these challenges. With motion up the chain, where data of higher order is inferred from lower-order data, the crucial question is whether privacy norms governing lower-order data are sufficient for the inferred higher-order data. While CI has a response to this question, a greater challenge comes from data primitives, such as digital impulses of mouse clicks, motion detectors, and bare GPS coordinates, because they appear to have no meaning. Absent a semantics, they escape CI’s privacy norms entirely.

Bot Detection on Social Networks Using Persistent Homology

Mathematical and Computational Applications ◽

10.3390/mca25030058 ◽

2020 ◽

Vol 25 (3) ◽

pp. 58

Author(s):

Minh Nguyen ◽

Mehmet Aktas ◽

Esra Akbas

Keyword(s):

Machine Learning ◽

Social Networks ◽

Social Media ◽

Persistent Homology ◽

Structural Features ◽

Higher Order ◽

Ego Networks ◽

Feature Extraction Method ◽

Topological Features ◽

Bot Detection

The growth of social media in recent years has contributed to an ever-increasing network of user data in every aspect of life. This volume of generated data is becoming a vital asset for the growth of companies and organizations as a powerful tool to gain insights and make crucial decisions. However, data is not always reliable, since primarily, it can be manipulated and disseminated from unreliable sources. In the field of social network analysis, this problem can be tackled by implementing machine learning models that can learn to classify between humans and bots, which are mostly harmful computer programs exploited to shape public opinions and circulate false information on social media. In this paper, we propose a novel topological feature extraction method for bot detection on social networks. We first create weighted ego networks of each user. We then encode the higher-order topological features of ego networks using persistent homology. Finally, we use these extracted features to train a machine learning model and use that model to classify users as bot vs. human. Our experimental results suggest that using the higher-order topological features coming from persistent homology is promising in bot detection and more effective than using classical graph-theoretic structural features.

Machine learning of higher order programs

Logical Foundations of Computer Science — Tver '92 - Lecture Notes in Computer Science ◽

10.1007/bfb0023859 ◽

2005 ◽

pp. 9-20 ◽

Cited By ~ 1

Author(s):

Ganesh Baliga ◽

John Case ◽

Sanjay Jain ◽

Mandayam Suraj

Keyword(s):

Machine Learning ◽

Higher Order

Localized Cognitive Functions

10.1093/med/9780192629760.003.0003 ◽

2013 ◽

Author(s):

John R. Hodges

Keyword(s):

Motor Control ◽

Cognitive Functions ◽

Right Hemisphere ◽

Higher Order ◽

Language Function ◽

Constructional Apraxia

Chapter 3 discusses aspects of normal and abnormal language function, followed by a brief description of disorders of calculation (acalculia) and of higher-order motor control (apraxia). The second half of the chapter deals with disturbed right hemisphere functions: neglect phenomena, dressing and constructional apraxia, and complex visuo-perceptual deficits (agnosias).

A Simple and Efficient Tensor Calculus

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v34i04.5881 ◽

2020 ◽

Vol 34 (04) ◽

pp. 4527-4534

Author(s):

Sören Laue ◽

Matthias Mitterreiter ◽

Joachim Giesen

Keyword(s):

Machine Learning ◽

Efficient Method ◽

Automatic Differentiation ◽

State Of The Art ◽

Higher Order ◽

Tensor Representation ◽

Tensor Calculus ◽

Correct Algorithm ◽

Previous State ◽

Derivatives Of

Computing derivatives of tensor expressions, also known as tensor calculus, is a fundamental task in machine learning. A key concern is the efficiency of evaluating the expressions and their derivatives that hinges on the representation of these expressions. Recently, an algorithm for computing higher order derivatives of tensor expressions like Jacobians or Hessians has been introduced that is a few orders of magnitude faster than previous state-of-the-art approaches. Unfortunately, the approach is based on Ricci notation and hence cannot be incorporated into automatic differentiation frameworks like TensorFlow, PyTorch, autograd, or JAX that use the simpler Einstein notation. This leaves two options, to either change the underlying tensor representation in these frameworks or to develop a new, provably correct algorithm based on Einstein notation. Obviously, the first option is impractical. Hence, we pursue the second option. Here, we show that using Ricci notation is not necessary for an efficient tensor calculus and develop an equally efficient method for the simpler Einstein notation. It turns out that turning to Einstein notation enables further improvements that lead to even better efficiency.

Higher Order Geometric Theory of Information and Heat Based on Poly-Symplectic Geometry of Souriau Lie Groups Thermodynamics and Their Contextures: The Bedrock for Lie Group Machine Learning

Entropy ◽

10.3390/e20110840 ◽

2018 ◽

Vol 20 (11) ◽

pp. 840 ◽

Cited By ~ 6

Author(s):

Frédéric Barbaresco

Keyword(s):

Machine Learning ◽

Maximum Entropy ◽

Lie Groups ◽

Lie Group ◽

Data Analytics ◽

Geometric Theory ◽

Higher Order ◽

Small Data ◽

Algebra Structure ◽

Fisher Metric

We introduce poly-symplectic extension of Souriau Lie groups thermodynamics based on higher-order model of statistical physics introduced by Ingarden. This extended model could be used for small data analytics and machine learning on Lie groups. Souriau geometric theory of heat is well adapted to describe density of probability (maximum entropy Gibbs density) of data living on groups or on homogeneous manifolds. For small data analytics (rarified gases, sparse statistical surveys, …), the density of maximum entropy should consider higher order moments constraints (Gibbs density is not only defined by first moment but fluctuations request 2nd order and higher moments) as introduced by Ingarden. We use a poly-sympletic model introduced by Christian Günther, replacing the symplectic form by a vector-valued form. The poly-symplectic approach generalizes the Noether theorem, the existence of moment mappings, the Lie algebra structure of the space of currents, the (non-)equivariant cohomology and the classification of G-homogeneous systems. The formalism is covariant, i.e., no special coordinates or coordinate systems on the parameter space are used to construct the Hamiltonian equations. We underline the contextures of these models, and the process to build these generic structures. We also introduce a more synthetic Koszul definition of Fisher Metric, based on the Souriau model, that we name Souriau-Fisher metric. This Lie groups thermodynamics is the bedrock for Lie group machine learning providing a full covariant maximum entropy Gibbs density based on representation theory (symplectic structure of coadjoint orbits for Souriau non-equivariant model associated to a class of co-homology).

S16-3 Praxis movements: the role of the beta band in higher-order motor control

Clinical Neurophysiology ◽

10.1016/s1388-2457(10)60118-6 ◽

2010 ◽

Vol 121 ◽

pp. S29

Author(s):

L.A. Wheaton

Keyword(s):

Motor Control ◽

Higher Order ◽

Beta Band

Defining consciousness

Pragmatics & Cognition ◽

10.1075/pc.18.3.04gal ◽

2010 ◽

Vol 18 (3) ◽

pp. 561-569 ◽

Cited By ~ 5

Author(s):

Shaun Gallagher

Keyword(s):

Motor Control ◽

Higher Order ◽

Phenomenological Description ◽

Self Awareness ◽

Control Processes ◽

The Common

I review the problem of how to define consciousness. I suggest that rather than continuing that debate, we should turn to phenomenological description of experience to discover the common aspects of consciousness. In this way we can say that consciousness is characterized by intentionality, phenomenality, and non-reflective self-awareness. I explore this last characteristic in detail and I argue against higher-order representational theories of consciousness, with reference to blindsight and motor control processes.

The network underlying human higher-order motor control: Insights from machine learning-based lesion-behaviour mapping in apraxia

10.1101/536391 ◽

2019 ◽

Author(s):

Christoph Sperber ◽

Daniel Wiesen ◽

Georg Goldenberg ◽

Hans-Otto Karnath

Keyword(s):

Motor Control ◽

Temporal Cortex ◽

Neural Correlates ◽

Brain Network ◽

Higher Order ◽

Support Vector ◽

Precentral Gyrus ◽

Canonical Correlations ◽

Neurological Patients ◽

Left Hemisphere Stroke

AbstractNeurological patients with apraxia of pantomime provide us with a unique opportunity to study the neural correlates of higher-order motor function. Previous studies using lesion-behaviour mapping methods led to inconsistent anatomical results, reporting various lesion locations to induce this symptom. We hypothesised that the inconsistencies might arise from limitations of mass-univariate lesion-behaviour mapping approaches if our ability to pantomime the use of objects is organised in a brain network. Thus, we investigated apraxia of pantomime by using multivariate lesion behaviour mapping based both on support vector regression and sparse canonical correlations in a sample of 130 left-hemisphere stroke patients. Both multivariate methods identified multiple areas to underlie high-order motor control, including inferior parietal lobule, precentral gyrus, posterior parts of middle temporal cortex, and insula. Further, long association fibres were affected, such as the superior longitudinal fascicle, inferior occipito-frontal fascicle, uncinated fascicle, and superior occipito-frontal fascicle. The findings thus not only underline the benefits of multivariate lesion-behaviour mapping in brain networks, but they also uncovered that higher-order motor control indeed is based on a common anatomical network.