人類語(yǔ)言的內(nèi)容和結(jié)構(gòu)的分布式表示曾在上世紀(jì)80年代短暫的繁榮，但它很快就消失了，而在過去20年一直占據(jù)著繼續(xù)使用的語(yǔ)言明確表示，盡管在這些范疇的元素運(yùn)用概率或權(quán)重表示。然而，在過去的五年里已經(jīng)看到了復(fù)蘇，隨著高度成功的采用分布式向量空間表示，經(jīng)常在“神經(jīng)元”或“深度學(xué)習(xí)”模式的背景下。一個(gè)偉大的成功已經(jīng)發(fā)給字表示，我也會(huì)看一些我們近期工作的其他人更好地理解文字表述，以及他們?nèi)绾慰梢员徽J(rèn)為是全球矩陣分解的，而且，更類似于傳統(tǒng)文學(xué)。但是，計(jì)算機(jī)在職研究生需要的不僅僅是字表示：我們需要明白，是做出來的話較大的語(yǔ)言單位，已經(jīng)少得多解決的問題。我將討論使用分布式表示的樹結(jié)構(gòu)遞歸神經(jīng)網(wǎng)絡(luò)模型，展示他們?nèi)绾文軌蛱峁┱Z(yǔ)義相似性，情緒，句法解析結(jié)構(gòu)和邏輯蘊(yùn)涵復(fù)雜的語(yǔ)言模型。

　　克里斯托弗·曼寧是斯坦福大學(xué)計(jì)算機(jī)科學(xué)和語(yǔ)言學(xué)的教授。他的研究目標(biāo)是可以智能化的過程，理解和生成人類的語(yǔ)言材料的計(jì)算機(jī)。曼寧專注于機(jī)器學(xué)習(xí)方法來解決計(jì)算語(yǔ)言學(xué)的問題，包括句法分析，計(jì)算語(yǔ)義和語(yǔ)用學(xué)，文字推理，機(jī)器翻譯和深層次的學(xué)習(xí)NLP。他是ACM研究員，一AAAI研究員，以及ACL研究員，并與人合著領(lǐng)先統(tǒng)計(jì)自然語(yǔ)言處理和信息教科書。

    原文：Distributed representations of human language content and structure had a brief boom in the 1980s, but it quickly faded, and the past 20 years have been dominated by continued use of categorical representations of language, despite the use of probabilities or weights over elements of these categorical representations. However, the last five years have seen a resurgence, with highly successful use of distributed vector space representations, often in the context of "neural" or "deep learning" models.  One great success has been distributed word representations, and I will look at some of our recent work and that of others on better understanding word representations and how they can be thought of as global matrix factorizations, much more similar to the traditional literature. But we need more than just word representations: We need to understand the larger linguistic units that are made out of words, a problem which has been much less addressed. I will discuss the use of distributed representations in tree-structured recursive neural network models, showing how they can provide sophisticated linguistic models of semantic similarity, sentiment, syntactic parse structure, and logical entailment.