《卫报》展评 | 将思维转化为艺术的意识培育者,皮埃尔 · 于热在蛇形画廊的展览

2018-12-21 19:33 浏览:163 A+ | A-


皮埃尔 · 于热个展“Uumwelt”现场 / 蛇形画廊,伦敦


皮埃尔 · 于热

Uumwelt

2018.10.03 - 2019.02.10

蛇形画廊,伦敦


“当艺术家并非唯一的操作者和意图制造者时;当出现的结果并非必然取决于艺术家,而是一整个超出人类掌控的智能、生物体以及非生物体系统时;当眼前的情形不具有持续性,不指向任何人,且保持中立时,展览的仪式性也许将会自我生发并呈现。”

— 皮埃尔 · 于热与汉斯 · 乌尔里希 · 奥布里斯特的对谈,2018


皮埃尔 · 于热(Pierre Huyghe)1962年生于巴黎,目前主要在纽约生活与创作。他被看作是当今最杰出的观念艺术家之一,其创作往往基于推测模式。他创造复杂的环境系统,生物与非生物、真实与象征性的各种因素在其中相互依存、自我组织且共同进化,最终形成一个不稳定的动态网络。


他的个展“Uumwelt”(即德语的“环境”)正在伦敦蛇形画廊举办。展厅空间成为一个具有渗透性和偶发性的环境,将不同形式的认知、新兴智能、生物繁殖以及本能行为容纳其中。


在展览之际,记者海蒂 · 朱达(Hettie Judah)为英国《卫报》采访了艺术家,文章如下。


皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © Ola Ri

皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © Ola Rindal,图片提供:艺术家及蛇形画廊


意识的培育者:将思维转化为艺术的机器

海蒂 · 朱达,《卫报》,2018年10月2日


“我寻求的是某种能够自我创造的事物,它可以自行生发、演化或变异。” 皮埃尔 · 于热说道。这位法国艺术家坐在蛇形画廊的办公室中,低调地抽着电子烟,斑驳的阳光透过上端的窗户照在他脸上。他头发花白,双目炯炯有神。楼下的展厅中,新展览已经准备就绪。……


这位艺术家偏好的创作素材是具有自主性、不断变化、可自我调节的系统,例如蜂巢、分裂中的癌细胞、流感病毒、蚁穴以及水族馆中共生的生物群体等。它们的特性往往使作品的安装相当棘手。


当然,于热的创作形式还包括影像和雕塑。但在过去十年中,他最广为人知的是对艺术作品的再定义,将其转换为生态系统。他将不同类的事物 — 有些是生物,有些是非生物 — 置于同一个环境中,然后从中撤出,听任其自由演变。


皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © O

皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © Ola Rindal,图片提供:艺术家及蛇形画廊


于热在蛇形美术馆展出的新作品中,完全不见人类的痕迹。展览的主角也不是动物,而是人工智能(AI):五个大型LED屏幕向我们展示了人工智能试图分析人类大脑活动的过程。


今年1月,于热了解到日本京都的一个实验室的科学家正在研究深度图像重建(deep image reconstruction),以揭开人类大脑活动中复杂的视觉信息处理过程:在人工智能的帮助下,科学家可以得知你正在观看或想象的事物。这种读心术,是通过功能性核磁共振成像技术(functional MRI)来捕捉人类大脑活动,并收集数据。完成对大脑活动的分析后,人工智能在自身的大型图像数据库中进行比对,寻找相匹配的部分,从而判断出实验对象的意识中所出现的图像。


于热邀请一位实验参与者在脑中勾勒出一系列相互关联的图像,它们中有些直接来自图片,另一些则是基于艺术家的描述。在实验者逐一进行想象的同时,其大脑活动受到监测。在美术馆的屏幕上呈现的图像,就是人工智能检索图像库,寻找匹配结果的过程:我们观看的不是图像重建的最终结果,而是机器学习(machine learning)过程的一种视觉化呈现。关于最初的图像,于热不会透露任何信息:全靠观者自己去逐一破译它们有可能是什么,以及它们之间存在什么样的关联。


皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © O

皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © Ola Rindal,图片提供:艺术家及蛇形画廊


“当然,作为视觉艺术家,我深受这一新媒介的吸引”,于热表示。他运用机器学习来创造了一种不同寻常的“生态系统”,从一个“不受艺术家控制、充满复杂性与偶发性的构建世界”过渡到了“一个可想象的世界”。


在此,艺术家不再是操控一个物理的环境,而是,用他的话说,在某人的意识中“进行培育”。作品本身的自主性首先体现在于热和实验对象之间的感知差异 — 如果我请你想象一个苹果,你脑中出现的图像将有别于我的想象 — 其次则体现在人工智能进行的分析与图像重建过程。


于热对这一过程感到惊叹,同时也强调说深度图像重建“与现实或我们熟悉的图像都相去甚远”。我们在蛇形美术馆的屏幕上看到的,只是“由成千上万转瞬即逝的图像组成的幻想”。它们仿佛来自某种深度空间,或深深的噩梦:那是一个充满暗示的深坑,各种依稀可辨的痕迹在其中翻滚沸腾。


皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © O

皮埃尔 · 于热个展“Uumwelt”现场,蛇形画廊,伦敦 / © Ola Rindal,图片提供:艺术家及蛇形画廊


人工智能对图像的处理与人类的大脑大相径庭:前者显示的结果缺乏关联性逻辑。类人工智能(near-human intelligence)从细微处表现出的非人类特征,极为诡异。于热和我讨论了“恐怖谷理论”(uncanny valley):那里充斥着近似于人类,却不易辨别的实体。作为人类的我们会对其尤为警惕(恐怖片的怪物设计者就完全明白这一道理,因而最能引发恐惧感的怪物往往具备了更接近人类的属性)。


即使没有像亚历克斯 · 加兰(Alex Garland)的电影《机器姬》(Ex Machina)中那样逼真的机器人出现,于热的人工智能也能引发精神上的共鸣。对模式的识别并非机器学习所专有,我们人类也是如此,并且我们很难克制自己不去解读LED屏幕上闪现的图像。然而,仅仅是站在展厅中注视着屏幕,我们作为蛇形美术馆生态系统的一个构成部分,也会对整个环境产生影响。


皮埃尔 · 于热个展“Uumwelt”作品,蛇形画廊,伦敦 / © 京

皮埃尔 · 于热个展“Uumwelt”作品,蛇形画廊,伦敦 / © 京都大学神谷研究室及日本国际电气通信基础技术研究所,图片提供:艺术家及蛇形画廊


此外还有苍蝇。展厅中安装的一系列感应器,在捕捉人类参观者的行为、动作和密度之外,也会监控在空间中放入的5万只绿头蝇的行为模式。


虽然5万这个数字听起来像是《出埃及记》中的蝇灾,策展人瑞贝卡 · 勒温(Rebecca Lewin)告诉我,预期的效果更接近于一场“恼人的夏季虫患”。放置在展厅穹顶处的白糖将吸引苍蝇聚集于此。由于其繁殖离不开蛋白质,在整个展览期间还需要补充新的苍蝇进来。大量采购苍蝇也成为了勒温最不同寻常的策展挑战之一。人类参观者和绿头蝇的行为会触发展厅中的传感器,并传导至一个“神经网络”,进而改变屏幕上图像闪现的速度。


皮埃尔 · 于热个展“Uumwet”作品,蛇形画廊,伦敦 / © 京都

皮埃尔 · 于热个展“Uumwet”作品,蛇形画廊,伦敦 / © 京都大学神谷研究室及日本国际电气通信基础技术研究所,图片提供:艺术家及蛇形画廊


于热将自己定义为煽动者,而非创造者。这体现在作品中始终保持的艺术家自我弱化姿态。既然深度图像重建能根据大脑活动产生图像,也就表明了这种大脑活动和其产生的图像为我们所有人共有的属性:我们的想象力是普遍的。在视觉艺术领域,这样的想法令人心生谦逊甚至尴尬。假如并非意识使我们显得与众不同呢,于热问道。


在蛇形美术馆,一个围绕着机器学习构建的系统,以可控的方式呈现了我们的世界的缩影。在其中,算法预测我们的行为,追踪所有的一切,从搜索关键词到我们的行迹。与此同时,在该作品中,在作品之外,我们的行为始终会对人类与其它有机生物体共享的环境产生深刻的影响。


皮埃尔 · 于热个展“Uumwelt”作品,蛇形画廊,伦敦 / © 京

皮埃尔 · 于热个展“Uumwelt”作品,蛇形画廊,伦敦 / © 京都大学神谷研究室及日本国际电气通信基础技术研究所,图片提供:艺术家及蛇形画廊






Pierre Huyghe

Uumwelt

3 October 2018 - 10 Febrary 2019

The Serpentine Gallery, London


"When what is made is not necessarily due to the artist as the only operator, the only one generating intentions and that instead it's an ensemble of intelligences, of entities biotic or abiotic, beyond human reach, and that the present situation has no duration, is not addressed to anyone, is indifferent, at that moment perhaps the ritual of the exhibition can self-present."

- Pierre Huyghe in conversation with Hans Ulrich Obrist, 2018


Born in Paris in 1962 and based in New York, Pierre Huyghe is considered as one of the world’s leading artists. Huyghe works on situations that are often based on speculative models. The environments he creates are complex systems in which interdependent agents, biotic and abiotic, real and symbolic are self-organising, co-evolving in a dynamic and unstable mesh.


He is presenting a major new exhibition Uumwelt, which means environment in German, at the Serpentine Gallery in London. For the exhibition, the gallery has become a porous and contingent environment, housing different forms of cognition, emerging intelligence, biological reproduction and instinctual behaviours.

 

On the occasion of the exhibition, Hettie Judah interviewed the artist and published in The Guardian the article below:


The Mind Gardener: The Machine that Turns your Thoughts into Art

Hettie Judah, The Guardian, 3 Oct 2018


“I was looking for something that could write itself, that could self-generate, evolve or mutate,” says Pierre Huyghe. Impeccably grizzled and sparkly eyed, the French artist sits vaping discreetly in dappled sunshine beaming through the upper windows of the Serpentine Gallery in London. Downstairs, Huyghe's new exhibition is being installed... 


Autonomous, changing, self-regulatory systems such as colonies of bees, dividing cancer cells, flu virus, ants’ nests and groups of creatures cohabiting in an aquarium are this artist’s preferred raw materials. They tend to make his work tricky to install.


While he also makes film and sculpture, over the last decade Huyghe has become known for reimagining the artwork as ecosystem. He deposits groups of objects – some evidently alive, others not so – into an environment, then retires and allows nature to take its course.


Human, alas, will not make an appearance in Huyghe’s new work at the Serpentine. The star turn here is not animal, but artificial intelligence (AI), the processing of which we witness via five LED screens as it attempts to analyse events within the human brain.


In January, Huyghe learned of a laboratory in Kyoto in which scientists were working on deep image reconstruction to decode complex visual information from human brain activity: artificial intelligence that can go some way towardworking out what it is you’re looking at or visualising. This machine telepathy uses data harvested during functional magnetic resonance imaging (fMRI), which measures brain activity. After analysing this brain activity, the AI sorts through vast nets of example images already in its system looking for fragments of a match to help it determine what image the human subject was holding in their mind’s eye.


Huyghe worked with a man to whom he gave a set of interlinked images to visualise, some based on physical images, others that he invited him to imagine. The man pictured them one by one as his brain activity was monitored. What we see on the screens in the gallery is the AI moving through its exemplar images, sifting for echoes: we are looking not at the final result of the image reconstruction but a visible manifestation of the process of machine learning. Huyghe won’t disclose what the original images were: it’s up to us to decipher in turn what they might have been, and how they may be linked to one another.


“Of course as a visual artist it is very appealing – a new medium,” says Huyghe. With machine learning he has generated an “ecosystem” of a different kind, passing “from a world that is constructed – with all the complexity and contingencies that go with it to evolve without me – to a world that can be imagined”.


Rather than manipulating a physical environment, he is now, as he puts it, “gardening” in someone’s mind. The work asserts its autonomy first in the perceptual gap between Huyghe and his human subject – if I tell you to think of an apple, the apple you think of will not be the apple I think of – and then in the analysis and image construction undertaken by the AI.


While “very impressive”, Huyghe warns that deep image reconstruction “is far from something close to a reality or image that we know”. What we see on the Serpentine’s screens at any one time is “a chimera made out of thousands of past images and the present images that are just there for a fraction of a second”. They’re like something out of deep space or even deeper nightmare: a fleshy, suggestive sump, seething with traces of near recognisable detail.


AI does not process images as a human brain might: there’s no associative logic to what might emerge. Near-human intelligence that reveals itself to be subtly non-human is freaky. Huyghe and I discuss the idea of the “uncanny valley” – the zone occupied by entities that are nearly but not quite recognisably human. As humans we tend to find them particularly alarming. (Monster-makers for horror movies understand this perfectly; it’s why the most disturbing monsters tend to have attributes that recall Homo sapiens.)


Even without a 3D android as persuasive as the one in Alex Garland’s film Ex Machina, Huyghe’s AI will invite intellectual empathy. Pattern recognition isn’t only the preserve of machine learning: we humans do it, too, and we’ll find it hard to resist interpreting the images as they evolve on the LED screens. However, just by standing in the gallery looking at them, we, as components in the Serpentine gallery ecosystem, will have an impact on the overall environment.


Then there are the flies. Along with the behaviour, movements and density of human visitors, a set of sensors installed in the gallery will also monitor the behaviour patterns of 50,000 bluebottles introduced into the space.


Though 50,000 sounds akin to the fourth plague, the intended effect, curator Rebecca Lewin tells me, is something closer to an “irritating August infestation”. The flies will be fed sugar up in the rotunda, where they are expected to congregate. Unable to breed without the presence of protein, their stock will need replenishing over the course of the exhibition. Purchasing flies in bulk has been one of Lewin’s more unusual curatorial challenges. Human visitors and bluebottles alike will trigger sensors in the gallery, feeding into a “neural network” that will change the pace at which the images on the screens evolve.


Huyghe presents himself as an instigator, rather than author. That stance is in keeping with an ego-denting suggestion buried in the work – if deep image reconstruction can generate images from brain activity, it suggests that this brain activity and the images it generates are common to all of us: that something about our imagination is generic. In the field of visual art, that idea is humbling and not a little awkward. What if the mind is not what sets us apart from other people, asks Huyghe.


At the Serpentine, a system structured around machine learning offers a controlled microcosm of the world we have just stepped out of, where algorithms predict our behaviour, tracking everything from search words to our physical movements. Of equal, if not greater importance perhaps: within this work, just as outside it, human behaviour has a profound effect on an environment that we share with other living organisms.






即  将  展  出  I  UPCOMING EXHIBITIONS



N E W  Y O R K

RICHARD DEACON

HOUSE AND GARDEN

JANUARY 9 – FEBRUARY 16, 2019


P A R I S

AMAR KANWAR

SUCH A MORNING

JANUARY 12 – MARCH 7, 2019


L O N D O N

DRAWINGS

JANUARY 24 – FEBRUARY 23, 2019



《卫报》展评 | 将思维转化为艺术的意识培育者,皮埃尔 · 于热在蛇形画廊的展览,蛇形画廊,皮埃尔,卫报,展评,培育者,伦敦,人工智能,热个展,Uumwelt,生物

《卫报》展评 | 将思维转化为艺术的意识培育者,皮埃尔 · 于热在蛇形画廊的展览,蛇形画廊,皮埃尔,卫报,展评,培育者,伦敦,人工智能,热个展,Uumwelt,生物

《卫报》展评 | 将思维转化为艺术的意识培育者,皮埃尔 · 于热在蛇形画廊的展览,蛇形画廊,皮埃尔,卫报,展评,培育者,伦敦,人工智能,热个展,Uumwelt,生物





长按并识别二维码,关注画廊最新动态

如需转载,请注明出处

点击阅读原文,进入画廊官方网站




微信:MarianGoodmanGallery

官网:http://mariangoodman.com

脸书:Marian Goodman Gallery

Instagram:mariangoodmangallery  



阅读原文

*以上内容由所属艺客发布或授权发布,转载请注明出处。
本网站不承担相应版权归属责任,如有侵权可联系网站申诉或删除

发表评论
博文分类

艺讯订阅

联系客服
86-021-62666063
info@trueart.com

分享

×