Imitation of honey bees’ concept learning processes using Vector Symbolic Architectures

Kleyko, Denis; Osipov, Evgeny; Gayler, Ross W.; Khan, Asad I.; Dyer, Adrian G.

doi:10.1016/j.bica.2015.09.002

Cited by 35 publications

(19 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bees can thus only perform complex tasks when they are trained to do it explicitly (e.g., in case of conditional probabilities) or when their training protocol allows them to derive satisfactory responses based on their simpler learning (e.g., in the case of learning the premise pairs during the transitivity study). Nevertheless, these simpler solutions that bees use to solve complex visual problems could be inspirational for developing efficient artificial intelligence systems with reduced computational resources to do the same (70)(71)(72)(73), and provide insights into how larger brains automatically perform such tasks.…”

Section: Discussionmentioning

confidence: 99%

Different mechanisms underlie implicit visual statistical learning in honey bees and humans

Avarguès‐Weber

Finke

Nagy³

et al. 2020

Proc. Natl. Acad. Sci. U.S.A.

Self Cite

View full text Add to dashboard Cite

The ability of developing complex internal representations of the environment is considered a crucial antecedent to the emergence of humans’ higher cognitive functions. Yet it is an open question whether there is any fundamental difference in how humans and other good visual learner species naturally encode aspects of novel visual scenes. Using the same modified visual statistical learning paradigm and multielement stimuli, we investigated how human adults and honey bees (Apis mellifera) encode spontaneously, without dedicated training, various statistical properties of novel visual scenes. We found that, similarly to humans, honey bees automatically develop a complex internal representation of their visual environment that evolves with accumulation of new evidence even without a targeted reinforcement. In particular, with more experience, they shift from being sensitive to statistics of only elemental features of the scenes to relying on co-occurrence frequencies of elements while losing their sensitivity to elemental frequencies, but they never encode automatically the predictivity of elements. In contrast, humans involuntarily develop an internal representation that includes single-element and co-occurrence statistics, as well as information about the predictivity between elements. Importantly, capturing human visual learning results requires a probabilistic chunk-learning model, whereas a simple fragment-based memory-trace model that counts occurrence summary statistics is sufficient to replicate honey bees’ learning behavior. Thus, humans’ sophisticated encoding of sensory stimuli that provides intrinsic sensitivity to predictive information might be one of the fundamental prerequisites of developing higher cognitive abilities.

show abstract

Section: Discussionmentioning

confidence: 99%

Different mechanisms underlie implicit visual statistical learning in honey bees and humans

Avarguès‐Weber

Finke

Nagy³

et al. 2020

Proc. Natl. Acad. Sci. U.S.A.

Self Cite

View full text Add to dashboard Cite

show abstract

“…VSAs make use of additional operations on high-dimensional vectors. So far, VSAs have been applied in various fields including robotics [45], addressing catastrophic forgetting in deep neural networks [9], medical diagnosis [73], fault detection [29], analogy mapping [52], reinforcement learning [30], long-short term memory [11], text classification [31], and synthesis of finite state automata [49]. They have been used in combination with deep-learned descriptors before, e.g.…”

Section: Vector Symbolic Architecturesmentioning

confidence: 99%

Vector Semantic Representations as Descriptors for Visual Place Recognition

Neubert¹,

Schubert²,

Schlegel³

et al. 2021

Robotics: Science and Systems XVII

View full text Add to dashboard Cite

Place recognition is the task of recognizing the current scene from a database of known places. The currently dominant algorithmic paradigm is to use (deep learning based) holistic feature vectors to describe each place and use fast vector query methods to find matchings. We propose a novel type of image descriptor, Vector Semantic Representations (VSR), that encodes the spatial semantic layout from a semantic segmentation together with appearance properties in a, for example, 4,096 dimensional vector for place recognition. We leverage operations from the established class of Vector Symbolic Architectures to combine symbolic (e.g. class label) and numeric (e.g. feature map response) information in a common vector representation. We evaluate the proposed semantic descriptor on 13 standard mobile robotic place recognition datasets and compare to six descriptors from the literature. VSR is on par with the best compared descriptor (NetVLAD) in terms of mean average precision and superior in terms of recall and worst-case average precision. This makes the approach particularly interesting for candidate selection. For a more detailed investigation, we discuss and evaluate recall integrity as additional criterion. Further, we demonstrate that the semantic descriptor is particularly well suited for combination with existing appearance descriptors indicating that semantics provide complementary information for image matching.

show abstract

“…It is based on bio-inspired methods which are based on one of the mechanisms of brain activity. As it is known, relatively simple mental events involve a very large number of scattered neurons, similarly, the vector-character architecture uses a distributed representation of information, so that one logical entity is associated with a large number of codes [14][15][16][17][18][19][20]. There are many different types of vector-character architecture using different representations, for example, [12][13][14].…”

Section: Vector-character Architecturementioning

confidence: 99%

“…Cognitive abilities achieved using vector-character architectures were demonstrated in works [16,17] devoted to imitating learning systems of honey bees. In add ition, approaches to the use of vector-character architectures for solving progressive Raven matrices [18,19,21] were proposed.…”

Section: Vector-character Architecturementioning

confidence: 99%

Search for a substring of characters using the theory of non-deterministic finite automata and vector-character architecture

et al. 2020

View full text Add to dashboard Cite

The paper proposed an algorithm which purpose is searching for a substring of characters in a string. Principle of its operation is based on the theory of non-deterministic finite automata and vector-character architecture. It is able to provide the linear computational complexity of searching for a substring depending on the length of the searched string measured in the number of operations with hyperdimensional vectors when repeatedly searching for different strings in a target line. None of the existing algorithms has such a low level of computational complexity. The disadvantages of the proposed algorithm are the fact that the existing hardware implementations of computing systems for performing operations with hyperdimensional vectors require a large number of machine instructions, which reduces the gain from this algorithm. Despite this, in the future, it is possible to create a hardware implementation that can ensure the execution of operations with hyperdimensional vectors in one cycle, which will allow the proposed algorithm to be applied in practice.

show abstract

Imitation of honey bees’ concept learning processes using Vector Symbolic Architectures

Cited by 35 publications

References 37 publications

Different mechanisms underlie implicit visual statistical learning in honey bees and humans

Different mechanisms underlie implicit visual statistical learning in honey bees and humans

Vector Semantic Representations as Descriptors for Visual Place Recognition

Search for a substring of characters using the theory of non-deterministic finite automata and vector-character architecture

Contact Info

Product

Resources

About