ArticlePDF Available

Abstract and Figures

Can artificial intelligence systems exhibit superhuman performance, but in critical ways, lack the intelligence of even a single-celled organism? The answer is clearly ‘yes’ for narrow AI systems. Animals, plants, and even single-celled organisms learn to reliably avoid danger and move towards food. This is accomplished via a physical knowledgepreserving metamodel that autonomously generates useful models of the world. We posit that preserving the structure of knowledge is critical for higher intelligences that manage increasingly higher levels of abstraction, be they human or artificial. This is the key lesson learned from applying AGI subsystems to complex real-world problems that require continuous learning and adaptation. In this paper, we introduce the Deep Fusion Reasoning Engine (DFRE), which implements a knowledge-preserving metamodel and framework for constructing applied AGI systems. The DFRE metamodel exhibits some important fundamental knowledge preserving properties such as clear distinctions between symmetric and antisymmetric relations, and the ability to create a hierarchical knowledge representation that clearly delineates between levels of abstraction. The DFRE metamodel, which incorporates these capabilities, demonstrates how this approach benefits AGI in specific ways such as managing combinatorial explosion and enabling cumulative, distributed and federated learning. Our experiments show that the proposed framework achieves 94% accuracy on average on unsupervised object detection and recognition. This work is inspired by the state-of-the-art approaches to AGI, recent AGI-aspiring work, the granular computing community, as well as Alfred Korzybski’s general semantics.
Content may be subject to copyright.
A Metamodel and Framework For AGI
Hugo Latapie1and Ozkan Kilic2
1Cisco Systems
hlatapie@cisco.com
2Cisco Systems
okilic@cisco.com
Abstract. Can artificial intelligence systems exhibit superhuman per-
formance, but in critical ways, lack the intelligence of even a single-celled
organism? The answer is clearly ‘yes’ for narrow AI systems. Animals,
plants, and even single-celled organisms learn to reliably avoid danger
and move towards food. This is accomplished via a physical knowledge-
preserving metamodel that autonomously generates useful models of the
world. We posit that preserving the structure of knowledge is critical for
higher intelligences that manage increasingly higher levels of abstraction,
be they human or artificial. This is the key lesson learned from applying
AGI subsystems to complex real-world problems that require continuous
learning and adaptation. In this paper, we introduce the Deep Fusion
Reasoning Engine (DFRE), which implements a knowledge-preserving
metamodel and framework for constructing applied AGI systems. The
DFRE metamodel exhibits some important fundamental knowledge pre-
serving properties such as clear distinctions between symmetric and anti-
symmetric relations, and the ability to create a hierarchical knowledge
representation that clearly delineates between levels of abstraction. The
DFRE metamodel, which incorporates these capabilities, demonstrates
how this approach benefits AGI in specific ways such as managing com-
binatorial explosion and enabling cumulative, distributed and federated
learning. Our experiments show that the proposed framework achieves
94% accuracy on average on unsupervised object detection and recog-
nition. This work is inspired by the state-of-the-art approaches to AGI,
recent AGI-aspiring work, the granular computing community, as well as
Alfred Korzybski’s general semantics.
Keywords: Artificial Intelligence ·AI ·AGI ·General Semantics ·Levels
of Abstraction ·Neurosymbolic ·Cognitive Architecture
1 Introduction
After over 30 years of intense effort, the AGI community has developed the
theoretical underpinnings for AGI and affiliated working software systems [32,
33]. While achieving human-level AGI is arguably years to decades away, some
of the currently available AGI subsystems are ready to be incorporated into
non-profit and for-profit products and services. Some of the most promising
AGI systems we have encountered are OpenNARS [33, 34], OpenCog [11, 12],
and AERA [28]. We are collaborating with all three teams and have developed
video analytics and smart city applications that leverage both OpenCog and
OpenNARS [14].
After several years of applying these AGI technologies to complex real-world
problems in IoT, networking, and security at scale, we have encountered a few
stumbling blocks largely related to real-time performance on large datasets and
cumulative learning [29]. In order to move from successful proofs of concept
and demos to scalable products, we have developed the DFRE metamodel and
associated DFRE framework, which is the focus of this paper. We have used
this metamodel and framework to bring together a wide array of technologies
ranging from machine learning, deep learning, and probabilistic programming to
the reasoning engines operating under the Assumption of Insufficient Knowledge
and Resources (AIKR) [32]. As discussed below, we believe the initial results
are promising: Data show a dramatic increase in system accuracy, ability to
generalize, resource utilization, and real-time performance.
The DFRE metamodel and framework are based on the idea that knowledge
is structure —More specifically, that knowledge is hierarchical structure, where
there are levels corresponding to levels of abstraction. The DFRE metamodel
refers to how knowledge is hierarchically structured while a model refers to
knowledge stored in a manner complying to the DFRE metamodel. It is based
on non Aristotelian, non-elementalistic, system thinking. The backbone of its
hierarchical structure is based on difference, aka antisymmetric relations, while
the leaves of this structure are based on symmetric relations. As in figure 1, even
a simple amoeba has to keep track of distinctions and similarities because not
preserving symmetric and antisymmetric relations is fatally important.
Fig. 1. Amoeba distinguishing between distinctions and similarities.
Korzybski [21] dedicated the majority of his professional life to analyzing and
studying the nature of this hierarchical structure. While it is well beyond the
2
scope of this paper to discuss the details of his analysis, our initial focus was to
incorporate these fundamental principles.
K1 – the core framework of knowledge is based on anti-symmetric relations
Spatial understanding: right/left/top/bottom
Temporal understanding: before/after
Corporal understanding: pain/satiation
Emotional understanding: happy/sad
Social understanding: friend/foe
Causal understanding: X causes Y
K2 – symmetric relations add further structure
A and B are friends
A is like B
K3 – knowledge is layered
Sensor data is on a different layer than high level symbolic information
Symbolic information B, which expands or provides context to symbolic
information A, is at a higher layer / level of abstraction
In the symbolic space, there are theoretically an infinite number of layers
i.e. it is always possible to refer to a symbol and expand upon it, thus
creating yet another level of abstraction
K4 – since knowledge is structure, any structure destroying operations such
as confusing levels of abstraction, treating an anti-symmetric relation as
symmetric, or vice-versa, can be classified as a knowledge corrupting and/or
destroying operation 1.
The DFRE Knowledge Graph (DFRE KG) groups information into four dif-
ferent levels as shown in Figure 2. These levels are labeled L0, L1, L2, and L*
and represent different levels of abstraction with L0 being closest to the raw
data collected from various sensors and external systems and L2 representing
the highest levels of abstraction typically obtained via mathematical methods
i.e. statistical learning and reasoning. Each layer can theoretically have infinitely
many sub-layers. L* represents the layer where the high-level goals and moti-
vations are stored, such as self-monitoring, self-adjusting and self-repair. There
is no global absolute level for a concept and all sub-levels in L2 are relative.
However, L0, L1, L2, L* are global concepts themselves. For example, an agent
can be instantiated to troubleshoot a problem, such as one related to object
recognition or computer networking. The framework promotes cognitive synergy
and metalearning, which refer to the use of different computational techniques
(e.g., probabilistic programming, Machine Learning/Deep Learning, and such)
to enrich its knowledge and address combinatorial explosion issues.
One advantage of the DFRE Framework is its integration of human domain
expertise, ontologies, prior learnings by the current DFRE KG-based system,
1In [general semantics] Korzybski argues that what is currently limiting humanity’s
advancement is the general lack of understanding of how our own abstracting mecha-
nisms work. He considered mankind to currently be in the childhood of humanity and
the day, if it should come, that humanity becomes generally aware of the metamodel,
is the day humanity enters into the “manhood of humanity” [20]
3
Fig. 2. DFRE Framework with four levels of abstraction.
and other similar systems, and additional sources of prior knowledge through
the middleware services. It provides a set of services that an Agent can utilize
as shown in Figure 3.
The Sensor Data Services are used to digitize any real world data, such as
video recordings. Similarly, the Data Structuring Services restructure data if
needed, e.g., rectifying an image. These two services are the basis for Image Pro-
cessing Services which provide a set of supervised and unsupervised algorithms
to detect objects, colors, lines, and other visual criteria. The Sensor Data Ana-
lytic Services analyze objects and create object boundaries enriched with local
properties such as an object’s size and coordinates, which create a 2D symbolic
representation of the world. Spatial Semantic Services then uses this representa-
tion to construct the initial knowledge graph that captures the spatial relations
of the object as a relational graph. Any (L2-) high-level reasoning is performed
on this knowledge graph.
Graph-based knowledge representation provides a system with the ability to
Capture the relations in the sub-symbolic world effectively in a world of
symbols,
Keep a fluid data structure independent of programming language in which
Agents running on different platforms can share and contribute,
Use algorithms based on the graph neural networks, which allow preservation
of topological dependency of information [25] on nodes.
Graph embedding [5, 13] is a technique used to represent graph nodes, edges
and sub-graphs in vector space that other machine learning algorithms can use.
4
Fig. 3. DFRE Framework.
Graph neural networks use graph embeddings to aggregate information from
graph structures in non-Euclidean ways. This allows DFRE Framework to use the
embeddings to learn from different data sources that are in the form of graphs,
such as Concept Net [26]. Despite its performance across different domains, the
graph neural networks suffer from scalability issues [41, 42] because calculating
the Laplacian matrix for all nodes in a large network may not be feasible. The
levels of abstraction and the focus of attention mechanisms used by an Agent
resolve this scalability issue in a systematic way.
All processes are fully orchestrated by the Agent that catalogues knowledge
by strictly preserving the structure while evolving new structures and levels of
abstraction in its knowledge graph because, for DFRE KG, knowledge is struc-
ture. Multiple agents can have not only individual knowledge graphs but also
a single knowledge graph on which all can cooperate and contribute. In other
words, multiple agents can work toward the same goal by sharing the same
knowledge graph synchronously or asynchronously. Different agents can have
partially or fully different knowledge graphs depending on their experience, and
share those entire graphs or their fragments through the communication chan-
nel provided by the DFRE Framework. Note that although the framework can
provide supervised machine learning algorithms if needed, the current IoT use
case is based on a retail store which requires unsupervised methods as explained
in the next section.
2 Experimental Results
DFRE Framework was previously tested in the smart city domain [14], in which
the system learns directly from experience with no initial training required (one-
shot) based on a fusion of sub-symbolic (object tracker) and symbolic (ontology
5
and reasoning).The current use case is based on object class recognition in a retail
store. Shelf space monitoring, inventory management and alerts for potential
stock shortages are crucial tasks for retailers who want to maintain an effective
supply chain management. In order to expedite and automate these processes,
and reduce both the requisite for human labor and the risk of human error,
several machine learning- and deep learning-based techniques have been utilized
[1, 7,9, 30]. Despite the high success rates, the main problems for such systems are
the requirements for a broad training set, including compiling images of the same
product with different lighting and from different angles, and retraining when a
new product is introduced or an existing product is visually updated. The current
use case does not demonstrate an artificial neural network-based learning. The
DFRE Framework has an artificial general intelligence-based approach to these
problems.
Fig. 4. Retail use case for DFRE Framework.
Before a reasoning engine operates on symbolic data within the context of
DFRE Framework, several services must be run, as shown in Figure 4. The flow
starts with a still image captured from a video camera that constantly records
the retail shelves by the Sensor Data Services as in Figure 4.a, which corresponds
to L0 in Figure 2. Next, the image is rectified by the Data Structuring Services
in Figure 4.b for better line detection by the Image Processing Services, as
displayed in Figure 4.c. The Image Processing Services in the retail case are
unsupervised algorithms used for color-based pixel clustering and line detection,
such as probabilistic Hough transform [19]. The Sensor Data Analytics Services
in Figure 4.d create the bounding boxes which represent the input in a 2D
world of rectangles, as shown in Figure 4.e. The sole aim of all these services
6
is to provide the DFRE KG with the best symbolic representation of the sub-
symbolic world in rectangles. Finally, the Spatial Semantics Services operate on
the rectangles to construct a knowledge graph, which preserves not only the
symbolic representation of the world, but also the structures within it in terms
of relations, as shown in Figure 4.f. This constitutes the L1 level abstraction in
the DFRE KG. L1 knowledge graph representation also recognizes and preserves
the attributes of each bounding box, such as the top-left xand ycoordinates, the
center’s coordinates, height,width,area and circumference. The relations used
for the current use case are inside,aligned,contains,above,below,on left of,on
right of,on top of,under and floating. Since DFRE KG enforces antisymmetrical
relations, the system does not know that aligned(a,b) means aligned(b,a), or on
left of and on right of are inverse relations unless such terms are input as expert
knowledge or are learned by the system through experience or simulations. As
far as the DFRE metamodel, the only innate relations are distinctions, which are
anti-symmetric and similarity relations and the rest is learned by experience.
7
Fig. 5. LoA for retail use case.
The system’s ultimate aim is to dynamically determine shelves,products and
unknown/others as illustrated in Figure 5 and to monitor the results with times-
tamps.
While L2 identifies only the concepts of shelf,products and unknowns and
the possible relations among them, a reasoning engine, NARS, a non-axiomatic
reasoning engine (see [33, 34, 36] for a detailed review and Narsese-NARS’s lan-
guage), is used to determine their extensions in L1. NARS is a powerful system
that operates with the assumption of insufficient knowledge and resources. It also
uses an evidence-based truth system in which there is no absolute knowledge.
This is useful in the retail use case scenario because the noise in L0 data causes
8
both overlapping regions and conflicting premises at L1. This noise results from
not only the projection of the 3D world input data into a 2D framework, but
also the unsupervised algorithms used by L1 services. The system has only four
rules for L2 level reasoning:
If a rectangle contains another rectangle that is not floating, the outer rect-
angle can be a shelf while the inner one can be a product.
If a rectangle is aligned with a shelf, it can be a shelf too.
If a rectangle is aligned with a product horizontally, it can be a product too.
If a floating rectangle is stacked on a product, it can be a product too.
Note that applying levels of abstraction gives DFRE Framework the power to
perform reasoning based on the expert knowledge in L2 level mostly independent
of L1 level knowledge. In other words, the system does not need to be trained
for different input and it is unsupervised in that sense. The system has a met-
alearning objective which is continuously attempting to improve it’s knowledge
representation. The current use case had 152 rectangles of various shapes and
locations, of which 107 were products, while 16 were shelves, and the remain-
ing 29 were other objects. When the knowledge graph in L1 is converted into
Narsese, 1,478 lines of premises that represent both the relations and attributes
were obtained and sent to the reasoner. Such a large amount of input with the
conflicting evidence caused the reasoning engine to perform poorly. Furthermore,
the symmetry and transitivity properties associated with the reasoner resulted in
scrambling the existing structure in the knowledge graph. Therefore, the DFRE
Framework employed a Focus of Attention (FoA) mechanism. The FoA creates
overlapping covers of knowledge graphs for the reasoner work on this limited
context. Later, the framework combines the results from the covers to finally
determine the intensional category. For example, when the FoA utilizes the rea-
soner on a region, a rectangle can be recognized as a shelf. However, when the
same rectangle is processed in another cover, it can be classified as a product.
The result with higher frequency and confidence will be the winner. The FoA
mechanism is inspired by the human visual attention system, which manages
input flow and recollects evidence as needed in case of a conflicting reasoning
result. A FoA mechanism can be based on objects attributes, such as color or
size, with awareness of proximity. For this use case, the FoA determined the
contexts by picking the largest non-empty rectangle and traversing its neighbors
based on their sizes in decreasing order.
DFRE Framework was tested 10 times with and without the FoA. The pre-
cision, recall and f-1 scores are exhibited in Table 1.
9
Table 1. DFRE Framework experimental results.
without FoA with FoA
category precision recall f1-score precision recall f1-score
product 80.70% 29.32% 52.88% 96.36% 99.07% 97.70%
shelf 8.82% 18.75% 12.00% 82.35% 87.50% 88.85%
other 36.61% 89.66% 52.00% 96.00% 82.76% 88.89%
overall accuracy 46.30% 94.73%
The results indicate that the FoA mechanism improves the success of our
AGI-based framework significantly by allowing the reasoner to utilize all of its
computing resources in a limited but controlled context. The results are accumu-
lated by the framework and the reasoner makes a final decision. This approach
not only allows us to perform reasoning on the intension sets of L1 knowledge,
which are retrieved through unsupervised methods, but also resolves the com-
binatorial explosion problem whose threshold depends on the limits of available
resources. In addition, one can easily extend this retail use case to include prior
knowledge of product types, and other visual objects, tables, chairs, people,
shelves, and such, as allowed by the DFRE KG.
3 Philosophical Implications
The nativism versus empiricism debate, which posits that some knowledge is in-
nate and some is learned through experience, was ascribed in the ancient world
by the Greek philosophers, including Plato and Epicurus. For the modern world,
Descartes is widely accepted as the pioneering philosopher working on the mind
as he furthered and reformulated the debate in the 17th century with new ar-
guments. Perception, memory and reasoning are three fundamental cognitive
faculties that enhance this debate by explicating the building blocks of natural
intelligence. We perceive the sub-symbolic world, and abstract it in memory, and
reason on this symbolic world representation. All three have concept learning and
categorization in the center of the human mind.
The process of concept learning and categorization continues to be an ac-
tive research topic related to the human mind, since it is essential to natural
intelligence [22] as well as cognitively inspired robotics research [3, 23]. It is
widely accepted that this process is based more on interactional properties and
relationships among agents, as well as between an agent and its environment
than objective features such as color, shape and size [18, 22]. This makes the
distinction of anti-symmetric and symmetric relations crucial in DFRE Frame-
work. DFRE Framework assumes that the levels of abstraction are part of innate
knowledge. In other words, an agent has L0, L1 and pre-existing L2 by default.
This constitutes a common a priori metamodel shared by all DFRE agents. Each
agent instantiated from the framework has the abstraction skill based on interac-
tional features and relationships. If the concepts in real life exist in interactional
systems, natural intelligence needs to capture these systems of interactions with
10
its own tools, such as abstraction. These tools should also be based on inter-
actional features by strictly preserving the distinction between symmetry and
anti-symmetry.
The mind is a system as well. Modern cognitive psychologists agree that con-
cepts and their relations in memory function as the fundamental data structures
to higher level system operations, such as problem solving, planning, reasoning
and language. Concepts are abstractions that have evolved from a conceptual
primitive. An ideal candidate for a conceptual primitive would be something
that is a step away from a sensorimotor experience [8], but is still an abstrac-
tion of experience [4]. A dog fails the mirror test but exhibits intelligence when
olfactory skills are needed to complete a task[16]. A baby’s mouthing behavior
is not only a requisite for developing oral skills but also for discovering the sur-
rounding environment through one of its expert sensorimotor skills related to
its survival. The baby is probably abstracting many objects into edible versus
inedible higher categories given its insufficient knowledge and resources. What
is astonishing about a natural intelligence system is that it does not need a
plethora of training input and experiments to learn the abstraction. It quickly
and automatically fits new information into an existing abstraction or evolves
it into a new one if needed. This is nature’s way of managing combinatorial
explosion. Objects and their interconnected relationships within the world can
be chaotic. Natural intelligence’s solution to this problem becomes its strength:
context. The concept of ‘sand’ has different abstractions depending on whether
it is on a beach, on a camera, or on leaves. An agent in these three different
contexts must abstract the sand in relation to its interaction with world in its
short-term memory. This cumulative set of experiences can later become part of
long-term memory, more specifically, episodic memory. DFRE Framework uses
a Focus of Attention (FoA) mechanism that provides the context while address-
ing the combinatorial explosion problem. The DFRE metamodel’s new way of
representing practically all knowledge as temporally evolving (i.e. time series)
can be viewed as the metamodel’s conceptual space. For example, the retail use
case given in Section 2 starts with a 3D world of pixels that is abstracted as
lines and rectangles in 2D. The framework produces spatial semantics using the
rectangles in the 2D world. Based on this situation, a few hundred rectangles
produce thousands of semantic relations, which present a combinatorial explo-
sion for most AGI reasoning engines. DFRE KG creates contexts, such candidate
shelves, for each scene, runs reasoners for each context, and merges knowledge
in an incremental way. This not only addresses the combinatorial explosion is-
sue, but also increases the success rate of reasoning, provided that the levels of
abstraction are computed properly [10].
Abstracting concepts in relation to their contexts also allows a natural intel-
ligence to perform mental experiments which is a crucial part of planning and
problem solving. DFRE Framework can integrate with various simulators, re-run
a previous example together with its context, and can alter what is known for
the purposes of experimentation to learn new knowledge, which is relationships
and interactions of concepts.
11
Several mathematical models and formal semantics [6, 2, 31, 37, 24] are pro-
posed to specify the meanings of real world objects as concept structures and
lattices. However, they are computationally expensive [17]. One way to over-
come this issue is with granular computing [40]. The extension of a concept can
be considered a granule and the intension of the concept is the description of
the granule. Assuming that concepts have granular common parts with varying
derivational and compositional stages and categorization, abstraction and ap-
proximation occur at multiple levels of granularity which has an important role
in human perception [15, 38, 39].
Having granular structures provides structured thinking, structured problem
solving, and structured information processing [40]. DFRE Framework has gran-
ular structures but emphasizes the preservation of structures in knowledge. Being
in the extension of a concept does not necessarily give the granular concept the
right to have similar relations and interactions of its intensional concept up to
certain degree or probability. Each level must preserve its inter-concept relation-
ships, and its symmetry or anti-symmetry. When a genuine problem that cannot
be solved by the current knowledge arises, it requires scrambling the structures
and running simulations on the new structures in order to provide an agent
with creativity. Note that this knowledge scrambling is performed in a separate
sandbox and the DFRE Framework ensures that the primary DFRE KG is not
corrupted by these creative, synthetic, “knowledge-scrambling” activities.
4 Conclusion
We have outlined key aspects of the DFRE Framework based on a simple idea
that knowledge is hierarchical structure. While this paper provides highlights
of one experiment in the visual domain employing an unsupervised approach,
we have also run similar experiments on time series and natural language data
with similar promising results. Based on the level of approvals we are able to
obtain, our plan is to release more details including mathematical formulations
of the DFRE Framework, source code, and other artifacts. At a minimum, prior
versions of the DFRE Framework utilized in the smartcity domain should soon
be released as open source.
We are always looking for collaborators (academic or commercial). Feel free
to contact the authors with your comments and suggestions.
References
1. Baz, I., Yoruk, E., Cetin, M.: Context-aware hybrid classification system for fine-
grained retail product recognition. In Image, Video, and Multidimensional Signal
Processing Workshop (IVMSP), 2016 IEEE 12th. IEEE, pp. 1-–5 (2016)
2. Belohl´avek, R.: Concept lattice and order in fuzzy logic. Annals of Pure Applied
Logic 128(1-3), 277–298 (2004)
3. Chella, A., Dindo, H., Infantino, I.: A cognitive framework for imitation learning,
in: The Social Mechanisms of Robot Programming by Demonstration, Robotics and
Autonomous Systems 54 (5), 403-–408 (2006)
12
4. Cohen, P.: Projections as Concepts. Computer Science Department Faculty Pub-
lication Series (194),https://scholarworks.umass.edu/cs faculty pubs/194. Last ac-
cessed 18 May 2020
5. Cui, P., Wang, X., Pei, J., Zhu, W.: A survey on network embedding. IEEE Trans-
actions on Knowledge and Data Engineering (2018).
6. untsch, I., Gediga, G.: Modal-style operators in qualitative data analysis. In: Pro-
ceedings of the 2002 IEEE International Conference on Data Mining, Washington,
D.C., pp. 155—162. Springer (2002)
7. Franco, A., Maltoni, D., Papi, S.: Grocery product detection and recognition. Expert
Systems with Applications 81 , 163—176 (2017).
8. ardenfors, P.: Conceptual Spaces: the Geometry of Thought. MIT Press, Cam-
bridge, Massachussetts (2000)
9. George, M., Floerkemeier, C.: Recognizing products: A perexemplar multi-label im-
age classification approach. In European Conference on Computer Vision. Springer,
pp.440-–455 (2014)
10. Gorban, Alexander N., and Ivan Yu Tyukin.: Blessing of dimensionality: math-
ematical foundations of the statistical physics of data. Philosophical Transactions
of the Royal Society A: Mathematical, Physical and Engineering Sciences 376.2118
(2018)
11. Goertzel, B.: OpenCogPrime: A Cognitive Synergy Based Architecture for General
Intelligence. International Conference on Cognitive Informatics, Hong Kong (2009)
12. Goertzel, B,. et al.: The CogPrime Architecture for Embodied Artificial General
Intelligence. Proceedings of IEEE Symposium on Human-Level AI, Singapore (2013)
13. Hamilton, W. L., Ying, R., Leskovec, J.: Representation learning on graphs: Meth-
ods and applications. IEEE Data(base) Engineering Bulletin, 40, 52-–74 (2017)
14. Hammer, P., Lofthouse, T., Fenoglio, E., and Latapie, H.: A reasoning based model
for anomaly detection in the SmartCity domain. In NARS Workshop at AGI-19,
Shenzhen, China. https://cis.temple.edu/tagit/events/papers/Hammer.pdf. Last
accessed 19 May 2020
15. Hobbs, J. R.: Granularity. In Proceedings of the 9th International Joint Conference
on Artificial Intelligence, pp. 432–435 (1985)
16. Horowitz, A.: Smelling themselves: dogs investigate their own odours longer when
modified in an olfactory mirror test. Behav. Proc. 143,17–24
17. Jinhai, L., Mei, C., Xu, W., Qian, Y.: 3 Concept learning via granular computing:
A cognitive viewpoint. Information Sciences 298, 447—467 (2015)
18. Johnson, M.: The Body in the Mind. University of Chicago Press, (1987)
19. Kiryati, N., Eldar, Y., Bruckstein, A. M. : A probabilistic Hough transform. Pattern
Recognition 244, 303–316 (1991)
20. Korzybski, A.: Manhood Of Humanity, The Science and Art of Human Engineering,
E. P. Dutton & Company, NY (1921)
21. Korzybski, A.: Science and Sanity: An Introduction to Non-Aristotelian Systems,
5th Edition. Institute of General Semantics, NY (1994)
22. Lakoff, G.: Women, Fire, and Dangerous Things. University of Chicago Press,
(1984)
23. Lieto, A., Chella, A., Frixione, M.: Conceptual spaces for cognitive architectures:
A lingua franca for different levels of representation, Biologically, Inspired Cognitive
Architectures 19, 1-–9 (2017)
24. Ma, J. M., Zhang, W. X., Leung, Y., Song, X. X.: Granular computing and dual
Galois connection. Information Sciences 177(23), 5365-–5377 (2007)
13
25. Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., Monfardini, G.: The graph
neural network model. IEEE Transactions on Neural Networks, 20(1), 61—80,
(2009)
26. Speer, R., Chin, J., Havasi, C.: ConceptNet 5.5: An Open Multilingual Graph of
General Knowledge. In proceedings of AAAI 31 (2017)
27. Th´orisson, K. R.: Integrated A.I. Systems. Minds & Machines, 17, 11–25 (2007)
28. Th´orisson, K. R.: A New Constructivist AI: From Manual Construction to Self-
Constructive Systems. In: P. Wang & B. Goertzel (Eds.), Theoretical Foundations
of Artificial General Intelligence, 4, pp. 145–171. Springer, New York (2012)
29. Th´orisson, K. R., Bieger, J., Li, X., Wang, P: Cumulative Learning. In: Proceed-
ings of 12th International Conference on Artificial General Intelligence (AGI-19),
Shenzhen, China, Aug. 6-9, 198–209 (2019)
30. Tonioni, A., Di Stefano, L.: Product recognition in store shelves as a sub-graph iso-
morphism problem. In International Conference on Image Analysis and Processing.
Springer, pp., 682-–693 (2017)
31. Wang, L. D., Liu, X. D.:Concept analysis via rough set and AFS algebra. Infor-
mation Sciences 178(21), 4125-–4137 (2008)
32. Wang, P.: Experience-grounded semantics: a theory for intelligent systems. Cogni-
tive System Research, 6, 282—302 (2005) doi:10.1016/j.cogsys.2004.08.003
33. Wang, P.: Rigid Flexibility: The Logic of Intelligence. Springer, Dordrecht (2006)
34. Wang, P.: Non-axiomatic logic (nal) specification (2010)
35. Wang, P.: On Defining Artificial Intelligence. Journal of Artificial General Intelli-
gence 10(2), 1–37, (2019)
36. Wang, P., Li, X., Hammer, P.: Self in NARS, an AGI System. Frontiers Robotics
and AI 5(20). doi: 10.3389/frobt.2018.00020
37. Wille, R.: Restructuring lattice theory: an approach based on hierarchies of con-
cepts. In: I. Rival (Ed.), Ordered Sets, pp. 445–470. Reidel, Dordrecht-Boston (1980)
38. Yao, Y.Y.: Information granulation and rough set approximation. International
Journal of Intelligent Systems 16, 87–104 (2001)
39. Yao, Y.Y.: Integrative levels of granularity. Bargiela A and Pedrycz W (eds.)
HumanCentric Information Processing Through Granular Modelling, pp. 31–47.
Springer, Berlin (2009)
40. Yao, Y.Y., Deng, X.F.: A Granular Computing Paradigm for Concept Learning.
In: Ramanna, S., Jain, L., Howlett, R.J. (Eds.), Emerging Paradigms in Machine
Learning, pp. 307–326. Springer, London (2012)
41. Ying, R., He, R., Chen, K., Eksombatchai, P., Hamilton, W. L., Leskovec, J.: Graph
convolutional neural networks for web-scale recommender systems. In Proceedings
of SIGKDD2018, (2018)
42. Zhou, J., Cui, G., Zhang, Z., Yang, C., Liu, Z., Sun, M.: Graph neural networks:
A review of methods and applications, 2018. https://arxiv.org/pdf/1812.08434.pdf.
Last accessed 24 May 2020
14
ResearchGate has not been able to resolve any citations for this publication.
Conference Paper
Full-text available
Using a proprietary visual scene object tracker and the Open-NARS reasoning system we demonstrate how to predict and detect various anomaly classes. The approach combines an object tracker with a base ontology and the OpenNARS reasoning system to learn to classify scene regions based on accumulating evidence from typical entity class (tracked object) behaviours. The system can autonomously satisfy goals related to anomaly detection and respond to user Q&A in real time. The system learns directly from experience with no initial training required (one-shot). The solution is a fusion of sub-symbolic (object tracker) and symbolic (ontology and reasoning).
Article
Full-text available
This article systematically analyzes the problem of defining “artificial intelligence.” It starts by pointing out that a definition influences the path of the research, then establishes four criteria of a good working definition of a notion: being similar to its common usage, drawing a sharp boundary, leading to fruitful research, and as simple as possible. According to these criteria, the representative definitions in the field are analyzed. A new definition is proposed, according to it intelligence means “adaptation with insufficient knowledge and resources.” The implications of this definition are discussed, and it is compared with the other definitions. It is claimed that this definition sheds light on the solution of many existing problems and sets a sound foundation for the field.
Article
Full-text available
This article describes and discusses the self-related mechanisms of a general-purpose intelligent system, NARS. This system is designed to be adaptive and to work with insufficient knowledge and resources. The system’s various cognitive functions are uniformly carried out by a central reasoning-learning process following a “non-axiomatic” logic. This logic captures the regularities of human empirical reasoning, where all beliefs are revisable according to evidence, and the meaning of concepts are grounded in the system’s experience. NARS perceives its internal environment basically in the same way as how it perceives its external environment although the sensors involved are completely different. Consequently, its self-knowledge is mostly acquired and constructive, while being incomplete and subjective. Similarly, self-control in NARS is realized using mental operations, which supplement and adjust the automatic inference control routine. It is argued that a general-purpose intelligent system needs the notion of a “self,” and the related knowledge and functions are developed gradually according to the system’s experience. Such a mechanism has been implemented in NARS in a preliminary form.
Article
Full-text available
The concentrations of measure phenomena were discovered as the mathematical background to statistical mechanics at the end of the nineteenth/beginning of the twentieth century and have been explored in mathematics ever since. At the beginning of the twenty-first century, it became clear that the proper utilization of these phenomena in machine learning might transform the curse of dimensionality into the blessing of dimensionality . This paper summarizes recently discovered phenomena of measure concentration which drastically simplify some machine learning problems in high dimension, and allow us to correct legacy artificial intelligence systems. The classical concentration of measure theorems state that i.i.d. random points are concentrated in a thin layer near a surface (a sphere or equators of a sphere, an average or median-level set of energy or another Lipschitz function, etc.). The new stochastic separation theorems describe the thin structure of these thin layers: the random points are not only concentrated in a thin layer but are all linearly separable from the rest of the set, even for exponentially large random sets. The linear functionals for separation of points can be selected in the form of the linear Fisher’s discriminant. All artificial intelligence systems make errors. Non-destructive correction requires separation of the situations (samples) with errors from the samples corresponding to correct behaviour by a simple and robust classifier. The stochastic separation theorems provide us with such classifiers and determine a non-iterative (one-shot) procedure for their construction. This article is part of the theme issue ‘Hilbert’s sixth problem’.
Article
Full-text available
Network embedding assigns nodes in a network to low-dimensional representations and effectively preserves the network structure. Recently, a significant amount of progresses have been made toward this emerging network analysis paradigm. In this survey, we focus on categorizing and then reviewing the current development on network embedding methods, and point out its future research directions. We first summarize the motivation of network embedding. We discuss the classical graph embedding algorithms and their relationship with network embedding. Afterwards and primarily, we provide a comprehensive overview of a large number of network embedding methods in a systematic manner, covering the structure- and property-preserving network embedding methods, the network embedding methods with side information and the advanced information preserving network embedding methods. Moreover, several evaluation approaches for network embedding and some useful online resources, including the network data sets and softwares, are reviewed, too. Finally, we discuss the framework of exploiting these network embedding methods to build an effective system and point out some potential future directions.
Conference Paper
Full-text available
The arrangement of products in store shelves is carefully planned to maximize sales and keep customers happy. Verifying compliance of real shelves to the ideal layout, however, is a costly task currently routinely performed by the store personnel. In this paper, we propose a computer vision pipeline to recognize products on shelves and verify compliance to the planned layout. We deploy local invariant features together with a novel formulation of the product recognition problem as a sub-graph isomorphism between the items appearing in the given image and the ideal layout. This allows for auto-localizing the given image within aisles of the store and improves recognition dramatically.
Article
Full-text available
Machine learning on graphs is an important and ubiquitous task with applications ranging from drug design to friendship recommendation in social networks. The primary challenge in this domain is finding a way to represent, or encode, graph structure so that it can be easily exploited by machine learning models. Traditionally, machine learning approaches relied on user-defined heuristics to extract features encoding structural information about a graph (e.g., degree statistics or kernel functions). However, recent years have seen a surge in approaches that automatically learn to encode graph structure into low-dimensional embeddings, using techniques based on deep learning and nonlinear dimensionality reduction. Here we provide a conceptual review of key advancements in this area of representation learning on graphs, including matrix factorization-based methods, random-walk based algorithms, and graph convolutional networks. We review methods to embed individual nodes as well as approaches to embed entire (sub)graphs. In doing so, we develop a unified framework to describe these recent approaches, and we highlight a number of important applications and directions for future work.
Article
While domestic dogs, Canis familiaris, have been found to be skillful at social cognitive tasks and even some meta-cognitive tasks, they have not passed the test of mirror self-recognition (MSR). Acknowledging the motivational and sensory challenges that might hinder performance, even before the question of self-recognition is broached, this study creates and enacts a novel design extrapolated from the species' natural behaviour. Given dogs' use of olfactory signals in communication, this experiment presents dogs with various canisters for approach and investigation. Each holds an odorous stimulus: in the critical test, either an “olfactory mirror” of the subject − the dog's own urine − or one in which the odour stimulus is modified. By looking at subjects' investigation times of each canister, it is shown that dogs distinguish between the olfactory “image” of themselves when modified: investigating their own odour for longer when it had an additional odour accompanying it than when it did not. Such behaviour implies a recognition of the odour as being of or from “themselves." The ecological validity of this odour presentation is examined by presenting to the subjects odours of other known or unknown dogs: dogs spend longer investigating the odour of other dogs than their own odour. Finally, in a second experiment, subjects spent longer with the modified stimulus than with the modified odour by itself, indicating that novelty alone does not explain the dogs' behavior. This study translates the MSR study for a species whose primary sensory modality is olfaction, and finds both that natural sniffing behaviour can be replicated in the lab and that dogs show more investigative interest in their own odours when modified.
Article
Object detection and recognition are challenging computer vision tasks receiving great attention due to the large number of applications. This work focuses on the detection/recognition of products in supermarket shelves; this framework has a number of practical applications such as providing additional product/price information to the user or guiding visually impaired customers during shopping. The automatic creation of planograms (i.e., actual layout of products on shelves) is also useful for commercial analysis and management of large stores.