A Computational Foundation for the Study of Cognition

( David J Chalmers ) 서울대학교 인지과학연구소 2011 Journal of Cognitive Science Vol.12 No.4

Computation is central to the foundations of modern cognitive science, but its role is controversial. Questions about computation abound: What is it for a physical system to implement a computation? Is computation sufficient for thought? What is the role of computation in a theory of cognition? What is the relation between different sorts of computational theory, such as connectionism and symbolic computation? In this paper I develop a systematic framework that addresses all of these questions. Justifying the role of computation requires analysis of implementation, the nexus between abstract computations and concrete physical systems. I give such an analysis, based on the idea that a system implements a computation if the causal structure of the system mirrors the formal structure of the computation. This account can be used to justify the central commitments of artificial intelligence and computational cognitive science: the thesis of computational sufficiency, which holds that the right kind of computational structure suffices for the possession of a mind, and the thesis of computational explanation, which holds that computation provides a general framework for the explanation of cognitive processes. The theses are consequences of the facts that (a) computation can specify general patterns of causal organization, and (b) mentality is an organizational invariant, rooted in such patterns. Along the way I answer various challenges to the computationalist position, such as those put forward by Searle. I close by advocating a kind of minimal computationalism, compatible with a very wide variety of empirical approaches to the mind. This allows computation to serve as a true foundation for cognitive science.

Defending the Semantic Conception of Computation in Cognitive Science

( Gerard O Brien ) 서울대학교 인지과학연구소 2011 Journal of Cognitive Science Vol.12 No.4

Cognitive science is founded on the conjecture that natural intelligence can be explained in terms of computation. Yet, notoriously, there is no consensus among philosophers of cognitive science as to how computation should be characterised. While there are subtle differences between the various accounts of computation found in the literature, the largest fracture exists between those that unpack computation in semantic terms (and hence view computation as the processing of representations) and those, such as that defended by Chalmers (2011), that cleave towards a purely syntactic formulation (and hence view computation in terms of abstract functional organisation). It will be the main contention of this paper that this dispute arises because contemporary computer science is an amalgam of two different historical traditions, each of which has developed its own proprietary conception of computation. Once these historical trajectories have been properly delineated, and the motivations behind the associated conceptions of computation revealed, it becomes a little clearer which should form the foundation for cognitive science.

Modal Considerations on Information Processing and Computation in the Nervous System

( Abel Wajnerman Paz ) 서울대학교 인지과학연구소 2013 Journal of Cognitive Science Vol.14 No.3

We can characterize computationalism very generally as a complex thesis with two main parts: the thesis that the brain (or the nervous system) is a computational system and the thesis that neural computation explains cognition. As Piccinini and Bahar (2012) point out, over the last six decades, computationalism has been the mainstream theory of cognition. Nevertheless, there is still substantial debate about which type of computation explains cognition, and computationalism itself still remains controversial. My aim in this paper is to make two main contributions to the debate about the first subthesis of computationalism, i.e. that the brain is a computational system. First, I want to offer an accurate elucidation of the notion relevant for understanding computationalism (the notion of computation) and clarify the relation between computation and information as well as the relations between both computation and information processing and the nervous system. Second, I want to argue against a peculiar form of computationalism: the thesis that neural processes are constitutively computational in some sense; that neural processes cannot be realized by a system that is not in some sense computational. I will call this thesis “modal computationalism.” In particular, I want to argue that neural processing can be realized by a system that is not a sui generis computer (i. e., a computing system that is neither digital nor analog) and by a system that is not a generic computer (a computer in the most general sense: one that includes digital, analog, and any other kind of computation). Actual neural processing is presumed to be computational in these two senses (Piccinini and Bahar 2012). I will argue that, even if this is true, neural processing can be realized by a computing system that is not of the same kind as those that perform actual neural processing and even by a system that is not computational at all.

A Problem for the Mechanistic Account of Computation

( Sabrina Haimovici ) 서울대학교 인지과학연구소 2013 Journal of Cognitive Science Vol.14 No.2

The mechanistic account of computation proposes that computational explanation is mechanistic, i.e. it explains the behavior and capacities of mechanisms in terms of their components and the functions and organization of those components. According to this account, computing systems are mechanisms that perform computations, that is, they process vehicles according to rules that are sensitive to certain vehicle properties. Despite the emphasis mechanists place on the structural description of mechanisms` components and activities, the description of computations does not rely on the structural individuation of components and activities, but only on their functional individuation. Concrete computations and their vehicles are not described in terms of detailed structural properties, but are instead characterized in a medium-independent way, that is, independently of the physical medium that implements the computation. Thus, the same specific computation can be implemented in different physical media, allowing the multiple realizability of computational systems. In this paper, I argue that the mechanistic account faces a dilemma. If computations and computational systems are individuated in functional terms, then computational explanations are elliptic mechanistic explanations, or mechanism sketches. But, according to mechanists, mechanism sketches are incomplete and explanatorily weak. Alternatively, for the computational explanation to satisfy the criteria for a good mechanistic explanation, we need a new way to individuate computations based on structural properties. However, as a result of this, multiple real- izability will no longer be possible for computational systems.

A Study on a Simple Algorithm for Parallel Computation of a Grid-Based One-Dimensional Distributed Rainfall-Runoff Model

최윤석,신문주,김경탁 대한토목학회 2020 KSCE Journal of Civil Engineering Vol.24 No.2

This paper presents an algorithm that can efficiently simulate a grid-based one-dimensional distributed rainfall-runoff model by performing parallel computations using flow accumulation values for individual grid cells, which are calculated through an eight flow direction method. This parallel computation algorithm uses information about flow accumulation to automatically find parallel computation target grid cells within the overall area and perform parallel computation on the grid by unit. The Microsoft .NET Parallel class was used to apply and evaluate the parallel computation algorithm independently on two machines. The results showed that the time reduction effect of parallel computation differed for each target domain, because flow accumulation values varied depending on the domain. Parallel computation reduced computation time by around 40% to 78% in virtual domains and around 63% in the real domain compared to sequential computation. The results of this study can be utilized to reduce the computation time of distributed models.

RNA nanotechnology for computer design and <i>in vivo</i> computation

Qiu, Meikang,Khisamutdinov, Emil,Zhao, Zhengyi,Pan, Cheryl,Choi, Jeong-Woo,Leontis, Neocles B.,Guo, Peixuan Royal Society 2013 Philosophical transactions. Series A, Mathematical Vol.371 No.2000

<P>Molecular-scale computing has been explored since 1989 owing to the foreseeable limitation of Moore's law for silicon-based computation devices. With the potential of massive parallelism, low energy consumption and capability of working <I>in vivo</I>, molecular-scale computing promises a new computational paradigm. Inspired by the concepts from the electronic computer, DNA computing has realized basic Boolean functions and has progressed into multi-layered circuits. Recently, RNA nanotechnology has emerged as an alternative approach. Owing to the newly discovered thermodynamic stability of a special RNA motif (Shu <I>et al.</I> 2011 <I>Nat. Nanotechnol.</I> <B>6</B>, 658–667 (doi:10.1038/nnano.2011.105)), RNA nanoparticles are emerging as another promising medium for nanodevice and nanomedicine as well as molecular-scale computing. Like DNA, RNA sequences can be designed to form desired secondary structures in a straightforward manner, but RNA is structurally more versatile and more thermodynamically stable owing to its non-canonical base-pairing, tertiary interactions and base-stacking property. A 90-nucleotide RNA can exhibit 4<SUP>90</SUP> nanostructures, and its loops and tertiary architecture can serve as a mounting dovetail that eliminates the need for external linking dowels. Its enzymatic and fluorogenic activity creates diversity in computational design. Varieties of small RNA can work cooperatively, synergistically or antagonistically to carry out computational logic circuits. The riboswitch and enzymatic ribozyme activities and its special <I>in vivo</I> attributes offer a great potential for <I>in vivo</I> computation. Unique features in transcription, termination, self-assembly, self-processing and acid resistance enable <I>in vivo</I> production of RNA nanoparticles that harbour various regulators for intracellular manipulation. With all these advantages, RNA computation is promising, but it is still in its infancy. Many challenges still exist. Collaborations between RNA nanotechnologists and computer scientists are necessary to advance this nascent technology.</P>

Efficient design optimization of complex system through an integrated interface using symbolic computation

Kim, Hansu,Kim, Shinyu,Kim, Taekyun,Lee, Tae Hee,Ryu, Namhee,Kwon, Kihan,Min, Seungjae Elsevier 2018 Advances in engineering software Vol.126 No.-

<P><B>Abstract</B></P> <P>A complex system is composed of several subsystems and numerous lower level components. It is inefficient to design the complex system at the system level by using high-fidelity models. A system level approach using a language-based model is required to design such a highly complex system before implementing a detailed design. However, as commercial process integration and design optimization software packages focus on detailed design, which uses the high-fidelity models, it is difficult to perform the design optimization of a highly complex system, such as a combat vehicle. Moreover, as commercial optimization software packages use numerical computation, the gradient calculation cost can increase, and the matrix computation process can be inefficient in terms of optimization time. Therefore, in this study, an integrated interface for efficient design optimization was developed to focus on concept design using a MODELICA language-based model. Additionally, the design variable screening by using analysis of variance, surrogate modeling through sequential design of experiments, and symbolic computation were used to solve the aforementioned problems. These were applied to the design optimization of the combat vehicle system to demonstrate the effectiveness of the integrated interface and symbolic computation. In conclusion, a concept design utilizing a MODELICA language-based model was achieved, and the optimization time achieved by symbolic computation was largely reduced in comparison to the optimization time achieved by numerical computation.</P> <P><B>Highlights</B></P> <P> <UL> <LI> An integrated interface is developed to focus on concept design via Maple software. </LI> <LI> Outlier analysis, screening, kriging model are adopted to the integrated interface. </LI> <LI> Symbolic computation is implemented to reduce computational cost. </LI> <LI> Design optimization of combat vehicle system is applied to the integrated interface. </LI> <LI> Optimization time is reduced by the integrated interface and symbolic computation. </LI> </UL> </P>

수학학습장애아동과 수학학습부진아동의 암산 능력과 전략 비교

김자경 ( Ja Kyoung Kim ),김기주 ( Ki Joo Kim ) 대구대학교 한국특수교육문제연구소 2005 특수교육저널 : 이론과 실천 Vol.6 No.4

This paper investigates the competence and strategy use in mental computation of students with math disabilities (MD) and students with low achieving in math (LA). The subjects of this study are 24 elementary school students in 4th and 5th grades: half of them have MD and another half have LA. Both MD and LA students are matched in their sex, age, and grade. The Mental Computation Test(Reys & Reys, 1993) is employed to measure the competence of mental computation. The kinds and frequencies of mental computation strategies used in both groups are classified and analysed on the basis of the classification system for mental computation strategies. According to the results of this study, the MD students show significantly low scores than the LA students in their mental computation tests. However, the kinds and frequencies of mental strategies adopted by the two groups are not significantly different. Despite their similarities in strategy use, the LA students most frequently use the ``mental form of written algorithm`` strategy while the MA students most frequently use the ``worked from the left`` strategy. This paper shows that MD can be interpreted as ‘delay’ not as ‘deficit’ and also that instruction about mental computation need to be conducted by the special education teachers in the special education classroom.

엣지 컴퓨팅 환경에서의 딥 러닝 연산 오프로딩의 병렬 최적화

신광용,문수묵 한국정보과학회 2022 정보과학회논문지 Vol.49 No.3

Computation offloading to edge servers has been proposed as a solution to performing computation-intensive deep learning applications on devices with low hardware capabilities. However, the deep learning model has to be uploaded to the edge server before computation offloading is possible, a non-trivial assumption in the edge server environment. Incremental offloading of neural networks was proposed as a solution as it can simultaneously upload model and offload computation [1]. Although it reduced the model upload time required for computation offloading, it did not properly handle the model creation overhead, increasing the time required to upload the entire model. This work solves this problem by parallel optimization of model uploading and creation, decreasing the model upload time by up to 30% compared to the previous system. 컴퓨팅 자원이 부족한 디바이스에 연산량이 많은 딥 러닝 애플리케이션을 실행하기 위해 주변에 있는 서버에 연산을 오프로딩하는 엣지 컴퓨팅 기술이 제안되었다. 그러나 딥 러닝 연산을 오프로딩하기 위해서는 서버에 모델을 먼저 업로드해야 하는 단점이 있다. 이를 해결하기 위해 모델을 점진적으로 전송하는 동시에 서버가 클라이언트 연산을 대신 수행하는 점진적 오프로딩 시스템이 제안되었다[1]. 점진적 오프로딩 시스템은 오프로딩에 걸리는 시간을 크게 단축했으나, 모델 구축 시간을 고려하지 않아서 전체 모델 업로드 시간이 늘어나는 단점이 있었다. 본 논문은 모델 구축과 모델 업로드의 병렬 최적화를 통해 기존 시스템의 문제점을 해결해서 기존 시스템 대비 전체 모델 업로드 시간을 최대 30% 개선했다.

Opportunistic Function Computation for Wireless Sensor Networks

Sang-Woon Jeon,Bang Chul Jung IEEE 2016 IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS Vol.15 No.6

<P>Function computation over wireless sensor networks is investigated, where a set of K sensors observe their sensor readings and a fusion center wishes to learn a predefined function of the sensor readings via fading multiple access channels (MACs). In this paper, the arithmetic sum and type functions are considered since they can yield various fundamental sample statistics such as mean, variance, maximum, and minimum. We propose a novel opportunistic in-network computation (INC) framework in which a subset of sensors with large channel gains opportunistically participate in the transmission at each time, while all sensors simultaneously send their observations or only a single sensor sends its observation in the conventional schemes. We mathematically analyze the long-term average computation rate of the proposed INC, and prove that it achieves a nonvanishing computation rate even when the number of sensors K tends to infinity, which is in fact a significant improvement and the first theoretical result in fading MACs. Note that the computation rates of the conventional schemes become zero as K increases. We further show that a similar multiuser diversity gain is still achievable under delay constraints, which implies that the proposed INC is restricted to exploit a fixed and finite number of time slots (or fading instances) for the function computation.</P>