Igor Farkas

Comenius University, Department of Applied Informatics, Faculty Member

Followers

Following

Public Views

Address: http://cogsci.fmph.uniba.sk/~farkas

less

Steven Pinker

Harvard University

Don Ross

University College Cork

Armando Marques-Guedes

UNL - New University of Lisbon

Vittorio Gallese

Università degli Studi di Parma (Italy)

Adam M Croom

Case Western Reserve University

jesus tapia

(Benemérita) Universidad Autónoma de Puebla

John Sutton

Macquarie University

Eros Carvalho

Universidade Federal do Rio Grande do Sul

Jairus Grove

University of Hawaii at Manoa

Fabio Cuzzolin

Oxford Brookes University

InterestsView All (6)

Uploads

Papers by Igor Farkas

A connectionist model of associating proprioceptive and tactile modalities in a humanoid robot

2022 IEEE International Conference on Development and Learning (ICDL)

Exploration by self-supervised exploitation

arXiv (Cornell University), Feb 22, 2023

Reinforcement learning can solve decision-making problems and train an agent to behave in an envi... more Reinforcement learning can solve decision-making problems and train an agent to behave in an environment according to a predesigned reward function. However, such an approach becomes very problematic if the reward is too sparse and the agent does not come across the reward during the environmental exploration. The solution to such a problem may be in equipping the agent with an intrinsic motivation which will provide informed exploration during which the agent is likely to also encounter external reward. Novelty detection is one of the promising branches of intrinsic motivation research. We present Self-supervised Network Distillation (SND), a class of intrinsic motivation algorithms based on the distillation error as a novelty indicator where the target model is trained using self-supervised learning. We adapted three existing self-supervised methods for this purpose and experimentally tested them on a set of ten environments that are considered difficult to explore. The results show that our approach achieves faster growth and higher external reward for the same training time compared to the baseline models, which implies improved exploration in a very sparse reward environment. 1 , 2

Download

Regionálna analýza pomocou samoorganizujúcich sa máp

Politická ekonomie, 2000

Exploratory analysis of financial and economic data is being accepted as a very valuable and impo... more Exploratory analysis of financial and economic data is being accepted as a very valuable and important stage of solving of majority of financial and economic tasks. We introduce a short overview of neural network approaches and applications in the financial and economic domain. Self-organizing map (SOM) used provides a unique and integrating neural network approach for proper and effective initial analysis of multidimensional financial and economic data. We provide a brief sketch of the basic types of analysis by means of SOM and illustrate their use on analysis of Slovak regions in terms of their budget income and expenditure data.

Potlačenie motorických rytmov v mozgu pri pozorovaní pohybu

State debt management: Goals and tools

Je reprezentačný pluralizmus v kognitívnej vede nevyhnutný?

The origin of categorical representations of language in the brain

Self-Organizing Maps for Representing Structures

The State of the Art in Computational Intelligence, 2000

Abstract: We propose a novel neural network model for representing data structures. The model con... more Abstract: We propose a novel neural network model for representing data structures. The model consists of a hierarchy of Self-Organizing Maps (SOMs) equipped with leaky integrating units. Each of the maps is thus designed to represent sequences of data in a fashion resembling Barnsley&amp;#x27;s iterated function system. Each data structure is decomposed into a hierarchy of sequences where in all but the lowest levels a special symbol is substituted to represent corresponding subtrees. The advantage of this representation is ...

Samoorganizujúce sa mapy

fmfi-uk.hq.sk, 1997

Page 1. 1 7. Samoorganizujúce sa mapy Igor Farka1 7.1. Úvod V tejto kapitole sa budeme venova al... more

Modely horizontálneho regionálneho finančného vyrovnávania

Ekonomický časopis, 2000

Jednym z dôlezitych bodov návrhu Stratégie reformy verejnej správy v SR [1] je aj nové chápanie v... more Jednym z dôlezitych bodov návrhu Stratégie reformy verejnej správy v SR [1] je aj nové chápanie vstupnych a vystupnych financnych tokov v regióne. Snahou je co najprirodzenejsie definovat', resp. usmernit'tieto toky tak, aby vytvárali dostatocnú kladnú spätnu väzbu pre región a tak pôsobili stimulujúco na jeho d'alsí sociálno-ekonomicky rast. Je zrejmé, ze takáto strategia posilní vsetky regióny, ktoré majú aspoñ minimálny potencial ekonomickcho rastu. Na druhej strane môze vi est'к prehlbeniu sociálno-ekonomického ...

Regional analysis using self-organizing maps

Politicka Ekonomie, 2000

Recipročná kauzalita medzi mysľou a mozgom z neurovednej perspektívy

Kognice a umělý život V, 2005

Abstrakt: Vo svetle pribúdajúcich biologických poznatkov ponúkame neurovedný pohľad na odveký pro... more Abstrakt: Vo svetle pribúdajúcich biologických poznatkov ponúkame neurovedný pohľad na odveký problém vzťahu mysle a mozgu. Prezentovaný pohľad je redukcionistický v tom zmysle, že mentálne procesy chápeme ako dynamiku neurálnych procesov, a je súčasne dualistický v tom zmysle, že postuluje dve úrovne činnosti mozgu, ktoré sa vzájomne ovplyvňujú. Vzťah mozgu a mysle je transformovaný na vzťah neurónov k mozgu ako celku. Kauzalita zdola nahor sa prejavuje existenciou emergentných javov v mozgu, zatiaľ čo ...

Download

Konekcionizmus v náručí výpočtovej kognitívnej vedy

Connectionist model of action learning and naming

frontiersin.org, 2011

Home; About; Submit; Advertise & PR. Register; Login. Science: Cell Science: Biochemical Genetics... more

Konceptuálne východiská pre model stelesnenej mysle

Kvasnička, V., Kelemen, J., Pospíchal, J.: Modely mysle. Europa, Bratislava, 2008

Abstrakt. Zaoberám sa otázkou povahy udskej mysle ukotvenej v tele, a vlastnos ami, ktoré by mal ... more Abstrakt. Zaoberám sa otázkou povahy udskej mysle ukotvenej v tele, a vlastnos ami, ktoré by mal potenciálny výpo tový model ma, aby bol o najviac konzistentný s empirickými znalos ami o fungovaní udskej mysle/mozgu. Navrhovaný konceptuálny rámec pre takýto model sa opiera o konekcionistickú a dynamickú paradigmu v kognitívnej vede. Z filozofického h adiska je tento rámec platformou neredukcionistického fyzikalizmu.

Download

Lexical acquisition and developing semantic map

Neural Network World, 2003

In this paper, we describe a self-organizing neural network model that addresses the process of e... more In this paper, we describe a self-organizing neural network model that addresses the process of early lexical acquisition in young children. The growing lexicon is modeled by combined semantic word representations based on distributional statistics of words and on grounded semantic features of words. Changing semantic word representations are assumed to model the maturation of word meaning and serve as inputs to the growing semantic map. The model has been tested on a real child-directed parental language ...

Download

Konekcionistické modelovanie jazyka

Jazyk a kognícia. Kalligram, Bratislava, 2005

... Tento pokus mono povaovať za prvý krok smerom vpred k vysvetleniu temporálnych aspektov čít... more

Ubal

Proceedings of the 2019 2nd International Conference on Computational Intelligence and Intelligent Systems, 2019

Artificial neural networks, in particular the deep end-to-end architectures trained by error back... more Artificial neural networks, in particular the deep end-to-end architectures trained by error backpropagation (BP), are currently the topmost used learning systems. However, learning in such systems is only loosely inspired by the actual neural mechanisms. Algorithms based on local activation differences were designed as a biologically plausible alternative to BP. We propose Universal Bidirectional Activation-based Learning, a novel neural model derived from contrastive Hebbian learning. Similarly to what is assumed about learning in the brain, our model defines a single learning rule that can perform multiple ways of learning via special hyperparameters. Unlike others, our model consists of mutually dependent, yet separate weight matrices for different directions of activation propagation. We show that UBAL can learn different tasks (such as pattern retrieval, denoising, or classification) with different setups of the learning hyperparameters. We also demonstrate the performance of our algorithm on a machine learning benchmark (MNIST). The experimental results presented in this paper confirm that UBAL is comparable with a basic version BP-trained multilayer network and the related biologically-motivated models.

Safe Reinforcement Learning in a Simulated Robotic Arm

Lecture Notes in Computer Science, Dec 31, 2022

Ultimate Grounding of Abstract Concepts: A Graded Account

Journal of Cognition, 2022

A connectionist model of associating proprioceptive and tactile modalities in a humanoid robot

2022 IEEE International Conference on Development and Learning (ICDL)

Exploration by self-supervised exploitation

arXiv (Cornell University), Feb 22, 2023

Download

Regionálna analýza pomocou samoorganizujúcich sa máp

Politická ekonomie, 2000