Neuron overload and the juggling physician

生物神经形态计算（Neuromorphic Computing）是一种受生物大脑启发，采用类脑方式运行的计算形式。

其目的是借鉴生物神经网络的结构和功能特点，建立能模拟生物大脑处理信息方式的计算系统。

生物神经形态计算将生物神经网络的各个方面实现为电子电路上的模拟或数字副本，旨在理解大脑中学习和进步的动态过程，并将大脑灵感应用于通用认知计算。

与传统方法相比，生物神经形态计算具有能源效率、执行速度、对局部故障的鲁棒性和学习能力等优势。

大规模的神经形态机器基于两个互补的原则，即多核SpiNNaker机器和BrainScaleS物理模型机。

多核SpiNNaker 机器将100万个ARM处理器与一个基于数据包的网络连接起来，该网络针对神经动作电位的交换进行了优化。

而BrainScaleS物理模型机在20个硅晶片上实现了400万个神经元和10亿个突触的模拟电子模型。

这两台机器都集成到HBP实验室中，并为其配置、操作和数据分析提供完整的软件支持。

生物神经形态计算的主要优势在于其能源效率、执行速度、对局部故障的鲁棒性和学习能力。

与传统超级计算机上的模拟相比，生物神经形态计算可以以更快的速度运行，并且更接近于生物大脑的实际运行情况。

此外，生物神经形态计算还可以通过模拟生物大脑的学习过程来改进机器学习算法，从而提高其性能和适应性。

总之，生物神经形态计算是一种受生物大脑启发，采用类脑方式运行的计算形式，旨在理解大脑中学习和进步的动态过程，并将大脑灵感应用于通用认知计算。

它具有能源效率、执行速度、对局部故障的鲁棒性和学习能力等优势，是未来计算机科学领域的重要发展方向之一。

基于转铁蛋白受体(TfR1)的肿瘤与脑部疾病靶向治疗研究进展人转铁蛋白受体（TfR1)在不同组织器官中普遍表达，其主要功能是协助转铁蛋白在细胞和血脑屏障内外转运，维持细胞铁平衡。

在肿瘤细胞中以及血脑屏障中，TfR1的表达水平明显高于正常细胞组织，因此，TfR1被认为是肿瘤靶向治疗和脑部疾病靶向治疗的重要靶点。

基于TfR1靶向治疗的药物载体主要有转铁蛋白（Tf）、抗TfR1抗体、TfR1结合肽，这些生物大分子能与TfR1特异性结合，结合之后可以通过受体介导的跨胞转运机制进入细胞或穿过血脑屏障。

将小分子药与这些载体偶联可以促进许多亲水性的化疗药物或神经治疗药物进入肿瘤细胞或血脑屏障，而许多中枢神经治疗性大分子则主要通过融合蛋白的方式与抗TfR1抗体连接转运进入中枢神经系统。

Abstract：Human TfR1 was universally expressed in different tissues. The major function of TfR1 was to facilitate delivery of transferrin across cells and blood-brain barrier(BBB). As a result, iron homo-stasis was maintained. TfR1 was recognised as a critical target for tumor and brain disease therapy due to its over expression in tumor cells and BBB. In recent years, drug carriers based on TfR1 recognition were developed such as Transferrin (Tf）, anti-TfR1 antibody and TfR1 binding peptide. These carriers bind to TfR1 specifically and enter into cell or BBB through receptor mediated endocytosis. Chemicals conjugated with these carriers can be facilitated to enter into tumor cells and brain tissue. Therapeutic proteins can be engineered to fused with anti-TfR1 antibody and transported across BBB.Key words：TfR1; Tumor target therapy;Brain directed delivery1轉铁蛋白受体（TfR1)简介转铁蛋白受体（TfR1)是一种在不同组织和细胞系中普遍表达的糖蛋白。

综㊀㊀述㊀基金项目:国家自然科学基金(Ｎｏ.８１７７３７１４㊁８１５７３４１３㊁８１２７３４９７)㊀作者简介:刘思敏ꎬ女ꎬ研究方向:神经精神药理ꎬＥ－ｍａｉｌ:２４５９２１２６２３＠ｑｑ.ｃｏｍ㊀通信作者:洪浩ꎬ男ꎬ教授ꎬ博士生导师ꎬ研究方向:神经精神药理ꎬＴｅｌ:１３９５１６９６６８１ꎬＥ－ｍａｉｌ:ｈｏｎｇｈａｏ＠ｃｐｕ.ｅｄｕ.ｃｎ星形胶质细胞－神经元相互作用与抑郁症刘思敏ꎬ洪浩(中国药科大学药学院药理系ꎬ江苏南京２１１１９８)摘要:星形胶质细胞为神经元提供营养支持ꎬ参与信号传递ꎬ调节突触可塑性ꎬ对于维持中枢神经系统微环境稳态至关重要ꎮ近些年研究发现星形胶质细胞与神经元相互作用参与抑郁症发病过程ꎮ本文从能量代谢㊁营养支持以及突触可塑性层面ꎬ综述了近年来星形胶质细胞－神经元相互作用在抑郁症中的最新研究进展ꎬ以期将星形胶质细胞－神经元信号机制与抑郁症相结合ꎬ为抑郁症的预防和治疗提供新的理念和策略ꎮ关键词:星形胶质细胞ꎻ神经元ꎻ抑郁症ꎻ能量代谢ꎻ突触可塑性中图分类号:Ｒ７４９.４㊀文献标识码:Ａ㊀文章编号:２０９５－５３７５(２０２０)０３－０１５６－００５ｄｏｉ:１０.１３５０６/ｊ.ｃｎｋｉ.ｊｐｒ.２０２０.０３.００９Ｒｅｓｅａｒｃｈａｄｖａｎｃｅｓｏｆａｓｔｒｏｃｙｔｅ－ｎｅｕｒｏｎｉｎｔｅｒａｃｔｉｏｎｉｎｄｅｐｒｅｓｓｉｏｎＬＩＵＳｉｍｉｎꎬＨＯＮＧＨａｏ(ＤｅｐｏｒｔｍｅｎｔｏｆＰｈａｒｍａｃｏｌｏｇｙꎬＳｃｈｏｏｌｏｆＰｈａｒｍａｃｙꎬＣｈｉｎａＰｈａｒｍａｃｅｕｔｉｃａｌＵｎｉｖｅｒｓｉｔｙꎬＮａｎｊｉｎｇ２１１１９８ꎬＣｈｉｎａ)Ａｂｓｔｒａｃｔ:Ａｓｔｒｏｃｙｔｅｐｒｏｖｉｄｅｓｎｕｔｒｉｔｉｏｎｓｕｐｐｏｒｔｆｏｒｎｅｕｒｏｎｓꎬｐａｒｔｉｃｉｐａｔｅｓｉｎｓｉｇｎａｌｔｒａｎｓｍｉｓｓｉｏｎꎬａｎｄｒｅｇｕｌａｔｅｓｓｙｎａｐｔｉｃｐｌａｓｔｉｃｉｔｙꎬｗｈｉｃｈｉｓｅｓｓｅｎｔｉａｌｆｏｒｍａｉｎｔａｉｎｉｎｇｍｉｃｒｏｅｎｖｉｒｏｎｍｅｎｔａｌｈｏｍｅｏｓｔａｓｉｓｏｆｃｅｎｔｒａｌｎｅｒｖｏｕｓｓｙｓｔｅｍ.Ｉｎｒｅｃｅｎｔｙｅａｒｓꎬｉｔｗａｓｆｏｕｎｄｔｈａｔａｓｔｒｏｃｙｔｅ－ｎｅｕｒｏｎｓｉｎｔｅｒａｃｔｉｏｎｉｓｉｎｖｏｌｖｅｄｉｎｔｈｅｐａｔｈｏｇｅｎｅｓｉｓｏｆｄｅｐｒｅｓｓｉｏｎ.Ｉｎｔｈｉｓｐａｐｅｒꎬｔｈｅｌａｔｅｓｔｒｅｓｅａｒｃｈｐｒｏ￣ｇｒｅｓｓｏｆａｓｔｒｏｃｙｔｅ－ｎｅｕｒｏｎｉｎｔｅｒａｃｔｉｏｎｉｎｄｅｐｒｅｓｓｉｏｎｗａｓｒｅｖｉｅｗｅｄｆｒｏｍｔｈｅａｓｐｅｃｔｓｏｆｅｎｅｒｇｙｍｅｔａｂｏｌｉｓｍꎬｎｕｔｒｉｔｉｏｎｓｕｐｐｏｒｔａｎｄｓｙｎａｐｔｉｃｐｌａｓｔｉｃｉｔｙꎬｓｏａｓｔｏｃｏｍｂｉｎｅａｓｔｒｏｃｙｔｅ－ｎｅｕｒｏｎｓｉｇｎａｌｉｎｇｍｅｃｈａｎｉｓｍｗｉｔｈｄｅｐｒｅｓｓｉｏｎａｎｄｐｒｏｖｉｄｅｎｅｗｉｄｅａｓａｎｄｓｔｒａｔｅｇｉｅｓｆｏｒｔｈｅｐｒｅｖｅｎｔｉｏｎａｎｄｔｒｅａｔｍｅｎｔｏｆｄｅｐｒｅｓｓｉｏｎ.Ｋｅｙｗｏｒｄｓ:ＡｓｔｒｏｃｙｔｅꎻＮｅｕｒｏｎｓꎻＤｅｐｒｅｓｓｉｏｎꎻＥｎｅｒｇｙｍｅｔａｂｏｌｉｓｍꎻＳｙｎａｐｔｉｃｐｌａｓｔｉｃｉｔｙ㊀㊀抑郁症是常见的心境障碍性精神疾病ꎬ表现为情绪低落㊁抑郁性认知(无望㊁无助㊁无用)㊁快感缺失㊁意志活动减退㊁睡眠障碍等生物学症状ꎮ高患病率㊁高自杀风险和高致残率使得抑郁症成为全球主要的疾病负担[１]ꎮ在寻找潜在机制过程中ꎬ星形胶质细胞－神经元相互作用对抑郁症的影响逐渐引起重视ꎮ本文首先简述抑郁症中神经元㊁星形胶质细胞病理学研究ꎬ再从能量代谢㊁营养支持以及突触可塑性层面进行分析ꎬ最后总结星形胶质细胞－神经元相互作用在抑郁症中扮演的角色并展望未来面临的挑战ꎮ１㊀抑郁症中神经元病理学神经元作为中枢神经系统(ｃｅｎｔｒａｌｎｅｒｖｏｕｓｓｙｓｔｅｍꎬＣＮＳ)结构和功能的基本单位ꎬ其数量和形态变化与神经精神疾病密切相关ꎮ慢性应激会引起啮齿类动物神经活动改变ꎬ包括海马ＣＡ３锥体神经元萎缩ꎬ内侧前额叶皮质(ｍｅｄｉａｌｐｒｅｆｒｏｎｔａｌｃｏｒｔｅｘꎬｍＰＦＣ)Ⅱ/Ⅲ㊁Ⅴ层顶端树突数量和长度减少ꎬ腹侧苍白球小清蛋白中间神经元树突棘密度减少[２]ꎬ导致绝快感缺失及焦虑样行为ꎮ重度抑郁症(ｍａｊｏｒｄｅｐｒｅｓｓｉｖｅｄｉｓｏｒｄｅｒꎬＭＤＤ)患者海马神经纤维和齿状回成熟颗粒神经元减少[３]ꎬＰＦＣ锥体神经元和γ－氨基丁酸(ＧＡＢＡ)能神经元密度和大小降低[４]ꎮ２㊀抑郁症中星形胶质细胞病理学星形胶质细胞(ａｓｔｒｏｃｙｔｅꎬＡｓ)体积大且分支多㊁数量多且分布广泛ꎬ可动态接触突触㊁微血管系统ꎬ提供营养支持㊁促进突触形成和调节神经递质稳态并参与维护血脑屏障ꎬ对ＣＮＳ发育和功能至关重要ꎮ病理条件下Ａｓ表现为细胞肥大㊁胶质纤维酸性蛋白(ｇｌｉａｌｆｉｂｒｉｌｌａｒｙａｃｉｄｉｃｐｒｏｔｅｉｎꎬＧＦＡＰ)和中间纤丝增多ꎬ严重可致胶质疤痕增生[５]ꎮ大量证据表明ꎬＡｓ功能障碍是抑郁症发病的重要因素ꎮ抑郁症患者尸检发现大脑皮层和边缘脑区的Ａｓ密度显著降低[６]ꎬ下丘脑和尾状核ＧＦＡＰ表达显著减少[７]ꎮＡｓ逐渐成为包括抑郁症和精神分裂症在内的主要精神疾病病因学研究的热点ꎬ并可能成为药物作用的靶点ꎮ３　抑郁症中星形胶质细胞－神经元相互作用３.１㊀能量代谢和营养支持３.１.１㊀调节神经元能量和代谢㊀Ａｓ突起含大量葡萄糖转运体可直接从血液摄取葡萄糖ꎬ经无氧酵解产生乳酸为神经元供能ꎬ实现Ａｓ－神经元乳酸穿梭(ａｓｔｒｏｃｙｔｅ–ｎｅｕｒｏｎｌａｃｔａｔｅｓｈｕｔｔｌｅꎬＡＮＬＳ)[８]ꎮ神经元可影响ＡＮＬＳ中Ａｓ基因转录ꎬ从而诱导葡萄糖代谢和乳酸输出ꎬ控制Ａｓ代谢通量的稳态调节[９]ꎮ乳酸脱氢酶是ＡＮＬＳ的重要组成部分ꎬ抑制Ａｓ或神经元中乳酸脱氢酶可通过调节Ｎａ＋/Ｋ＋－ＡＴＰ通道导致神经元膜超极化ꎬ抑制神经元活性和突触传递[１０]ꎮ近年研究认为抑郁症发病伴随着脑内代谢平衡改变ꎬ包括神经递质和能量代谢紊乱ꎮＡｓ－神经元相互作用在ＣＮＳ能量代谢中发挥重要作用ꎬ一方面Ａｓ能量供应不足会减少树突分支ꎬ增加神经元易损性和抑郁易感性[１１]ꎮ另一方面神经元分泌的血管活性肠肽与Ａｓ表面血管活性肠肽受体结合ꎬ促进Ａｓ糖原分解[８]ꎬ还通过ｃＡＭＰ/ＰＫＡ途径调控Ａｓ葡萄糖代谢的ＣＲＥＢ依赖性增加和乳酸输出的升高[９]ꎬ神经元功能障碍可导致Ａｓ能量和乳酸供应不足ꎮ传统抗抑郁药氟西汀和帕罗西汀可减少Ａｓ糖原合成ꎬ促进葡萄糖代谢ꎬ从而促进神经功能恢复改善抑郁症状[１２]ꎮ以上结果提示Ａｓ－神经元功能障碍导致脑内能量代谢紊乱可能是导致抑郁症发生的重要因素之一ꎮ３.１.２㊀Ａｓ对神经元具有营养支持作用㊀神经营养因子包括神经生长因子㊁脑源性神经营养因子(ｂｒａｉｎ－ｄｅｒｉｖｅｄｎｅｕｒｏｔｒｏ￣ｐｈｉｃｆａｃｔｏｒꎬＢＤＮＦ)㊁胶质源性神经营养因子(ｇｌｉａｌ－ｄｅｒｉｖｅｄｎｅｕｒｏｔｒｏｐｈｉｃｆａｃｔｏｒꎬＧＤＮＦ)及成纤维母细胞生长因子等ꎬ生理条件下主要来源于Ａｓꎬ可促进神经生长发育㊁成熟分化ꎮＡｓ神经营养因子分泌障碍损害神经元功能ꎬ是目前公认的抑郁症病理发病机制之一ꎮ抑郁症患者海马ＢＤＮＦ㊁ＧＤＮＦ含量减少ꎬＰＦＣ纤维生长因子受体表达降低ꎬ且与海马神经再生减少有关[１２－１３]ꎮ传统抗抑郁药丙咪嗪通过ＰＫＡ－ＣＲＥＢ途径激活Ａｓꎬ促进ＢＤＮＦ和ＧＤＮＦ合成释放ꎬ发挥抗抑郁作用[１４]ꎮ速效抗抑郁药氯胺酮ꎬ雷帕替尼和东莨菪碱通过增加ＢＤＮＦ释放ꎬ产生原代皮层神经元保护作用ꎬ也可增加ｍＰＦＣ中树突棘数量和突触功能[１５]ꎮ以上结果表明ꎬ增加神经营养因子合成可能会是抑郁症的一种有效治疗策略ꎮ３.２㊀突触可塑性㊀突触可塑性(ｓｙｎａｐｔｉｃｐｌａｓｔｉｃｉｔｙ)是神经细胞持续活动导致突触特异性的结构和功能改变ꎬ与多种神经精神疾病的病理生理过程密切相关ꎮＡｓ通过调节神经递质(ＡＴＰ㊁谷氨酸㊁Ｄ－丝氨酸㊁γ－氨基丁酸等)影响突触强度和功能ꎬ其功能障碍可致突触异常ꎬ引起情绪调控㊁认知功能缺陷ꎬ导致抑郁症发生ꎮ３.２.１㊀ＡＴＰ㊀三磷酸腺苷(ａｄｅｎｏｓｉｎｅｔｒｉｐｈｏｓｐｈａｔｅꎬＡＴＰ)是脑内细胞重要的能量来源ꎬ也是广泛介导Ａｓ－神经元网络的重要信号分子ꎬ参与调节突触可塑性ꎮＡｓ释放的ＡＴＰ激活神经元上Ｐ２Ｘ７受体ꎬ从而增强α－氨基－３－羟基－５－甲基－４－异恶唑丙酸受体(ＡＭＰＡＲｓ)表达和微小兴奋性突触后突触电流[１６]ꎮ研究发现阻断Ａｓ释放ＡＴＰ可诱导小鼠海马神经棘数目减少ꎬ导致神经功能紊乱和抑郁样行为ꎬ外源性给予ＡＴＰ或内源性激活Ａｓ促进ＡＴＰ释放ꎬ可在一周内快速逆转抑郁样行为[１７]ꎮ氟西汀可促进Ａｓ通过囊泡核苷转运体释放ＡＴＰꎬ激活嘌呤Ｐ２Ｙ和腺苷Ａ２ｂ受体ꎬ增加ＢＤＮＦ表达ꎬ从而起到抗抑郁作用[１８]ꎮＡＴＰ介导的Ａｓ－神经元信号传递可能为抑郁症提供新的药物治疗思路ꎮ３.２.２㊀谷氨酸㊀谷氨酸(ｇｌｕｔａｍａｔｅꎬＧｌｕ)是脑内主要的兴奋性神经递质ꎬ主要通过Ａｓ特异性转运体(ｅｘｃｉｔａｔｏｒｙａｍｉｎｏａｃｉｄｔｒａｎｓｐｏｒｔｅｒｓꎬＥＡＡＴｓ)经谷氨酰胺合成酶(ｇｌｕｔａｍｉｎｅｓｙｎ￣ｔｈｅｔａｓｅꎬＧＳ)作用转变为谷氨酰胺(ｇｌｕｔａｍｉｎｅꎬＧｌｎ)ꎬ被神经元末梢再摄取ꎮ此外Ａｓ也可通过Ｃａ２＋依赖机制释放Ｇｌｕꎬ作用于突触代谢型谷氨酸受体(ｍｅｔａｂｏｌｉｃｇｌｕｔａｍａｔｅｒｅｃｅｐｔｏｒｓꎬｍＧｌｕＲｓ)调节突触传递[１９]ꎮ研究表明抑郁患者前扣带回及背外侧前额叶Ａｓ数量和ＥＡＡＴꎬＧＳ水平降低ꎬ调节Ｇｌｕ摄取和代谢的能力降低[２０]ꎮ慢性束缚应激ＣＩＳ模型小鼠ｍＰＦＣＧｌｎ水平和ＧＳ表达均下降ꎬ谷氨酸能神经元自发兴奋性突触后电流频率减少[２１]ꎮ乙酰左旋肉碱(ＬＡＣ)或乙酰－Ｎ－半胱氨酸(ＮＡＣ)可快速增加海马腹侧齿状回Ａｓ特异性谷氨酸交换子ｘＣＴ表达ꎬ激活ｍＧｌｕ２Ｒꎬ从而减少抑郁易感性并增加抗抑郁作用[２２]ꎮ神经元可通过Ｎｏｔｃｈ信号途径维持Ａｓ成熟和递质摄取ꎬ干扰神经元Ｎｏｔｃｈ信号转导会导致ＡｓＧｌｕ摄取和代谢障碍[９]ꎮ以上研究提示应激诱导Ａｓ损伤及Ｇｌｕ循环障碍ꎬ损伤神经元可塑性是引发抑郁症的关键因素ꎬ针对Ａｓ调节谷氨酸能神经传递将成为抑郁症治疗和新药开发策略之一ꎮ３.２.３㊀Ｄ－丝氨酸㊀Ｄ－丝氨酸(Ｄ－ｓｅｒｉｎｅ)主要由Ｓｅｒｉｎｅ消旋酶(ＳｅｒｉｎｅｒａｃｅｍａｓｅꎬＳＲ)将Ｌ－丝氨酸(Ｌ－ｓｅｒｉｎｅ)消旋而来ꎬ作用于Ｎ－甲基－Ｄ－天冬氨酸受体(Ｎ－ｍｅｔｈｙｌ－Ｄ－ａｓｐａｒｔａｔｅｒｅ￣ｃｅｐｔｏｒｓꎬＮＭＤＡＲｓ)ꎬＡｓ囊泡释放的Ｄ－ｓｅｒｉｎｅ可调节成年神经元ＮＭＤＡＲ介导的突触传递和树突成熟[２３]ꎮ关于Ｄ－ｓｅｒｉｎｅ主要来源至今仍具争议ꎬ起初研究认为主要由Ａｓ合成释放ꎬ逐渐有研究表明部分神经元也表达ＳＲꎮ原代Ａｓ培养发现Ｄ－ｓｅｒｉｎｅꎬＳＲ表达随着Ａ１型反应性Ａｓ的出现而明显增多ꎬ提示病理条件下Ａｓ过表达ＳＲ和Ｄ－ｓｅｒｉｎｅꎬ活化突触外ＮＭＤＡＲ导致神经毒性和突触功能障碍[２４]ꎮＤ－ｓｅｒｉｎｅ可促进ＳＲꎬ突触后密度蛋白ＰＳＤ９５ꎬＮＭＤＡＲ１相互作用ꎬ增强皮层和海马突触发育的稳定性[２５]ꎮ当神经元发生炎性损伤会引起Ａｓ细胞骨架波形蛋白及ＧＦＡＰ表达改变ꎬ促进Ａｓ释放Ｄ－ｓｅｒｉｎｅ[２６]ꎮ上述研究结果提示Ｄ－ｓｅｒｉｎｅ可作为影响抑郁症发病的重要因素之一ꎮ３.２.４㊀ＧＡＢＡ㊀γ－氨基丁酸(γ－ａｍｉｎｏｂｕｔｙｒｉｃａｃｉｄꎬＧＡＢＡ)是ＣＮＳ主要的抑制性神经递质ꎬ由谷氨酸脱羧酶合成ꎬ主要经ＧＡＢＡ转运体(ＧＡＢＡｔｒａｎｓｐｏｒｔｅｒｓꎬＧＡＴｓ)重摄取失活ꎮ大量研究表明ＧＡＢＡ水平异常及ＧＡＢＡ受体功能障碍与焦虑㊁抑郁等多种神经精神疾病相关ꎮＡｓ通过与海马ＧＡＢＡ能中间神经元相互作用动态调节ＧＡＢＡ抑制和Ｇｌｕ兴奋稳态ꎬ其中海马中间神经元ＧＡＢＡＡ受体参与调节突触抑制ꎬＡｓＧＡＢＡＢ受体参与调节突触增强作用ꎬ抑郁患者脑内出现Ａｓ功能障碍的同时ｍＰＦＣＧＡＢＡ能中间神经元减少[２７]ꎮ最近研究表明ꎬ低水平刺激ＧＡＢＡ能中间神经元可通过ＧＡＢＡＢ受体激活Ａｓꎬ从而释放Ｇｌｕ并激活ｍＧｌｕＲ１引起短暂突触效能增强ꎬ而高水平刺激诱导Ａｓ释放Ｇｌｕ激活ｍＧｌｕＲ１和ＡＴＰ激活Ａ１腺苷受体导致突触传递减弱[２８]ꎮ胶质细胞抑制剂Ｌ－ＡＡＡ可减少白介素１β诱导的下丘脑室旁核ＡｓＧＡＢＡ释放ꎬ改善焦虑样行为ꎬ表明选择性抑制下丘脑室旁核Ａｓ释放ＧＡＢＡ可能是治疗焦虑症和情感性精神障碍的有效治疗策略[２９]ꎮ上述研究提示Ａｓ－神经元之间存在复杂的ＧＡＢＡ传递ꎬ可能在焦虑或抑郁症状的病因学中起重要作用ꎮ３.２.５㊀突触重塑㊀Ａｓ－神经元相互作用可动态调节细胞形态和突触形成ꎮ神经元通过神经毒素与Ａｓ黏附蛋白(ＮＬ１㊁ＮＬ２㊁ＮＬ３)相互作用控制Ａｓ形态发生[３０]ꎬ而Ａｓ分泌血小板反应蛋白ꎬ结合神经元α２δ１受体刺激突触形成[３１]ꎮ应激导致的突触重塑伴随着抑郁症的发生ꎬｍＰＦＣ和海马突触缺失会导致焦虑和抑郁样症状ꎮ慢性不可预测应激(ｃｈｒｏｎｉｃｕｎ￣ｐｒｅｄｉｃｔａｂｌｅｓｔｒｅｓｓꎬＣＵＳ)模型小鼠中前额叶第５层锥体神经元树突长度与分支减少ꎬ突触棘数量及突触蛋白表达减少ꎬ兴奋性突触后电流降低[３２]ꎮＡｓ与神经元形态㊁突触结构的动态变化有着密切关联ꎬ参与调节突触重塑ꎮ双光子实时成像观察在体条件下Ａｓ可延长树突棘存活时间ꎬ促进突触形成与成熟[３３]ꎮ反复应激导致Ａｓ形态及分子功能的改变和神经元萎缩ꎬ提示应激条件下Ａｓ功能障碍诱导神经元萎缩及突触缺失可能是引发抑郁样行为的重要原因之一ꎮ３.２.６㊀间隙连接㊀间隙连接(ｇａｐｊｕｎｃｔｉｏｎꎬＧＪ)是在ＣＮＳ生理病理过程扮演重要角色的细胞间通道群ꎮＡｓ处于监测三突触活动的理想位置ꎬ且表达高水平的连接蛋白(ｃｏｎｎｅｘｉｎｓꎬＣｘｓ)ꎬ与神经元存在广泛的电位和代谢偶联ꎬ引导并促进神经元迁移分化ꎮ研究表明Ａｓ通过Ｃｘ４３和Ｃｘ３０共同调节神经传递ꎬ并摄取突触释放的Ｇｌｕ和Ｋ＋防止神经元过度激活ꎬＭＤＤ自杀患者背外腹侧前额叶皮质Ａｓ间隙连接蛋白Ｃｘ４３表达减少[３４]ꎮ氟西汀㊁度洛西汀和糖皮质激素受体拮抗剂米非司酮可逆转ＣＵＳ大鼠ＰＦＣＡｓ间隙连接功能障碍ꎬ并改善抑郁样行为[３５]ꎮ三环类抗抑郁药阿米替林可显著上调ＡｓＣｘ４３ｍＲＮＡ及蛋白表达水平ꎬ促进Ａｓ－神经元间隙连接通讯[３６]ꎮ此外ꎬＡｓ间隙连接功能障碍影响神经递质代谢ꎬ参与炎症反应ꎮ上述研究为抑郁症的发病机制提供了新的研究思路ꎮ３.２.７㊀神经再生㊀成年神经发生是神经可塑性的特殊形式ꎬ主要发生在侧脑室的室管膜下区(ＳＶＺ)和海马齿状回颗粒下区(ＳＧＺ)ꎬ产生向嗅球迁移的神经前体细胞和齿状回颗粒细胞ꎬ其中海马神经受损及再生障碍是抑郁症的重要发病机制之一ꎮＭＤＤ患者以及抑郁模型海马齿状回中新生神经元生成均减少ꎬ氟西汀以及非典型抗精神病药物奥氮平可增加海马和ＰＦＣ神经元再生[１８ꎬ３７]ꎮ给予脂多糖ＬＰＳ刺激可激活Ａｓꎬ诱导ＢＤＮＦ/ＴｒｋＢ和ＧＡＢＡＡＲ下调ꎬ损害海马新生神经元成熟ꎬ增加产后小鼠抑郁样行为[３８]ꎮ在小鼠海马Ａｓ过表达ＢＤＮＦ基因ꎬ可促进海马神经生成ꎬ明显改善新环境压抑进食试验中抑郁样行为[３９]ꎮ前体细胞产生的新生神经元属于放射状Ａｓꎬ提示Ａｓ参与神经再生可作为抗抑郁的潜在靶点ꎮ４　小结与展望抑郁症患者大脑存在神经递质紊乱㊁脑发育异常以及突触结构重塑等现象以及Ａｓ和功能的变化ꎮ随着抑郁症发病机制和药物靶点研究的不断深入ꎬ研发焦点逐渐从神经元转向ＡｓꎮＡｓ为神经元提供能量和营养支持ꎬ参与神经递质转运和代谢ꎬ积极调节突触可塑性ꎬ促进神经网络通讯交流ꎮ越来越多实验模型和临床证据表明Ａｓ－神经元相互作用是社会应激致抑郁症发病机制的一个重要因素(见图１)ꎮ然而仍存在许多问题和挑战ꎬ包括Ａｓ增生的生物标志物的进一步研究ꎬ各脑区Ａｓ不同状态与抑郁症不同阶段的相关性ꎬ单细胞转录组学等先进检测手段在抑郁症中的进一步应用ꎬ促进神经元图１　星形胶质细胞与神经元在抑郁症中的相互作用再生药物的筛选方法等ꎮ进一步阐明Ａｓ－神经元相互作用参与抑郁病理生理机制ꎬ可能成为抑郁症防治和新药开发的关键ꎮ参考文献:[１]㊀ＷＡＬＫＥＲＥＲꎬＭＣＧＥＥＲＥꎬＤＲＵＳＳＢＧ.Ｍｏｒｔａｌｉｔｙｉｎｍｅｎｔａｌｄｉｓ￣ｏｒｄｅｒｓａｎｄｇｌｏｂａｌｄｉｓｅａｓｅｂｕｒｄｅｎｉｍｐｌｉｃａｔｉｏｎｓ:ａｓｙｓｔｅｍａｔｉｃｒｅｖｉｅｗａｎｄｍｅｔａ－ａｎａｌｙｓｉｓ[Ｊ].ＪＡＭＡＰｓｙｃｈｉａｔｒｙꎬ２０１５ꎬ７２(４):３３４－３４１.[２]ＮＧＬꎬＨＵＡＮＧＹꎬＨＡＮＬꎬｅｔａｌ.Ｋｅｔａｍｉｎｅａｎｄｓｅｌｅｃｔｉｖｅａｃｔｉｖａｔｉｏｎｏｆｐａｒｖａｌｂｕｍｉｎｉｎｔｅｒｎｅｕｒｏｎｓｉｎｈｉｂｉｔｓｔｒｅｓｓ－ｉｎｄｕｃｅｄｄｅｎｄｒｉｔｉｃｓｐｉｎｅｅｌｉｍｉｎａｔｉｏｎ[Ｊ].ＴｒａｎｓｌＰｓｙｃｈｉａｔｒｙꎬ２０１８ꎬ８(１):２７２.[３]ＢＯＬＤＲＩＮＩＭꎬＳＡＮＴＩＡＧＯＡＮꎬＨＥＮＲꎬｅｔａｌ.Ｈｉｐｐｏｃａｍｐａｌｇｒａｎｕｌｅｎｅｕｒｏｎｎｕｍｂｅｒａｎｄｄｅｎｔａｔｅｇｙｒｕｓｖｏｌｕｍｅｉｎａｎｔｉｄｅｐｒｅｓｓａｎｔ－ｔｒｅａｔｅｄａｎｄｕｎｔｒｅａｔｅｄｍａｊｏｒｄｅｐｒｅｓｓｉｏｎ[Ｊ].Ｎｅｕｒｏｐｓｙｃｈｏｐｈａｒｍａ￣ｃｏｌｏｇｙꎬ２０１３ꎬ３８(６):１０６８－１０７７.[４]ＭÖＨＬＥＲＨ.ＴｈｅＧＡＢＡｓｙｓｔｅｍｉｎａｎｘｉｅｔｙａｎｄｄｅｐｒｅｓｓｉｏｎａｎｄｉｔｓｔｈｅｒａｐｅｕｔｉｃｐｏｔｅｎｔｉａｌ[Ｊ].Ｎｅｕｒｏｐｈａｒｍａｃｏｌｏｇｙꎬ２０１２ꎬ６２(１):４２－５３. [５]ＳＯＦＲＯＮＩＥＷＭＶꎬＶＩＮＴＥＲＳＨＶ.Ａｓｔｒｏｃｙｔｅｓ:ｂｉｏｌｏｇｙａｎｄｐａｔｈｏｌｏｇｙ[Ｊ].ＡｃｔａＮｅｕｒｏｐａｔｈｏｌꎬ２０１０ꎬ１１９(１):７－３５.[６]ＲＡＪＫＯＷＳＫＡＧ.Ｐｏｓｔｍｏｒｔｅｍｓｔｕｄｉｅｓｉｎｍｏｏｄｄｉｓｏｒｄｅｒｓｉｎｄｉｃａｔｅａｌ￣ｔｅｒｅｄｎｕｍｂｅｒｓｏｆｎｅｕｒｏｎｓａｎｄｇｌｉａｌｃｅｌｌｓ[Ｊ].ＢｉｏｌＰｓｙｃｈｉａｔｒｙꎬ２０００ꎬ４８(８):７６６－７７７.[７]ＴＯＲＲＥＳ－ＰＬＡＴＡＳＳＧꎬＮＡＧＹＣꎬＷＡＫＩＤＭꎬｅｔａｌ.Ｇｌｉａｌｆｉｂｒｉｌｌａｒｙａｃｉｄｉｃｐｒｏｔｅｉｎｉｓｄｉｆｆｅｒｅｎｔｉａｌｌｙｅｘｐｒｅｓｓｅｄａｃｒｏｓｓｃｏｒｔｉｃａｌａｎｄｓｕｂｃｏｒ￣ｔｉｃａｌｒｅｇｉｏｎｓｉｎｈｅａｌｔｈｙｂｒａｉｎｓａｎｄｄｏｗｎｒｅｇｕｌａｔｅｄｉｎｔｈｅｔｈａｌａｍｕｓａｎｄｃａｕｄａｔｅｎｕｃｌｅｕｓｏｆｄｅｐｒｅｓｓｅｄｓｕｉｃｉｄｅｓ[Ｊ].ＭｏｌＰｓｙｃｈｉａｔｒｙꎬ２０１６ꎬ２１(４):５０９－５１５.[８]ＭＡＧＩＳＴＲＥＴＴＩＰＪꎬＡＬＬＡＭＡＮＩ.Ａｃｅｌｌｕｌａｒｐｅｒｓｐｅｃｔｉｖｅｏｎｂｒａｉｎｅｎｅｒｇｙｍｅｔａｂｏｌｉｓｍａｎｄｆｕｎｃｔｉｏｎａｌｉｍａｇｉｎｇ[Ｊ].Ｎｅｕｒｏｎꎬ２０１５ꎬ８６(４):８８３－９０１.[９]ＨＡＳＥＬＰꎬＤＡＮＤＯＯꎬＪＩＷＡＪＩＺꎬｅｔａｌ.Ｎｅｕｒｏｎｓａｎｄｎｅｕｒｏｎａｌａｃｔｉｖｉｔｙｃｏｎｔｒｏｌｇｅｎｅｅｘｐｒｅｓｓｉｏｎｉｎａｓｔｒｏｃｙｔｅｓｔｏｒｅｇｕｌａｔｅｔｈｅｉｒｄｅｖｅｌ￣ｏｐｍｅｎｔａｎｄｍｅｔａｂｏｌｉｓｍ[Ｊ].ＮａｔＣｏｍｍｕｎꎬ２０１７(８):１５１３２. [１０]ＳＡＤＡＮꎬＬＥＥＳꎬＫＡＴＳＵＴꎬｅｔａｌ.Ｅｐｉｌｅｐｓｙｔｒｅａｔｍｅｎｔ.ＴａｒｇｅｔｉｎｇＬＤＨｅｎｚｙｍｅｓｗｉｔｈａｓｔｉｒｉｐｅｎｔｏｌａｎａｌｏｇｔｏｔｒｅａｔｅｐｉｌｅｐｓｙ[Ｊ].Ｓｃｉｅｎｃｅꎬ２０１５ꎬ３４７(６２２８):１３６２－１３６７.[１１]ＣＯＴＴＥＲＤＲꎬＰＡＲＩＡＮＴＥＣＭꎬＥＶＥＲＡＬＬＩＰ.Ｇｌｉａｌｃｅｌｌａｂｎｏｒ￣ｍａｌｉｔｉｅｓｉｎｍａｊｏｒｐｓｙｃｈｉａｔｒｉｃｄｉｓｏｒｄｅｒｓ:ｔｈｅｅｖｉｄｅｎｃｅａｎｄｉｍｐｌｉｃａｔｉｏｎｓ[Ｊ].ＢｒａｉｎＲｅｓＢｕｌｌꎬ２００１ꎬ５５(５):５８５－５９５.[１２]ＬＩＮＰＹꎬＴＳＥＮＧＰＴ.Ｄｅｃｒｅａｓｅｄｇｌｉａｌｃｅｌｌｌｉｎｅ－ｄｅｒｉｖｅｄｎｅｕｒｏｔｒｏｐｈｉｃｆａｃｔｏｒｌｅｖｅｌｓｉｎｐａｔｉｅｎｔｓｗｉｔｈｄｅｐｒｅｓｓｉｏｎ:Ａｍｅｔａ－ａｎａｌｙｔｉｃｓｔｕｄｙ[Ｊ].ＪＰｓｙｃｈｉａｔｒＲｅｓꎬ２０１５(６３):２０－２７.[１３]ＭＡＳＩＧꎬＢＲＯＶＥＤＡＮＩＰ.ＴｈｅＨｉｐｐｏｃａｍｐｕｓꎬＮｅｕｒｏｔｒｏｐｈｉｃＦａｃｔｏｒｓａｎｄＤｅｐｒｅｓｓｉｏｎＰｏｓｓｉｂｌｅＩｍｐｌｉｃａｔｉｏｎｓｆｏｒｔｈｅＰｈａｒｍａｃｏｔｈｅｒａｐｙｏｆＤｅｐｒｅｓｓｉｏｎ[Ｊ].ＣＮＳＤｒｕｇｓꎬ２０１１ꎬ２５(１１):９１３－９３１.[１４]ＴＡＫＡＮＯＫꎬＹＡＭＡＳＡＫＩＨꎬＫＡＷＡＢＥＫꎬｅｔａｌ.Ｉｍｉｐｒａｍｉｎｅｉｎｄｕｃｅｓｂｒａｉｎ－ｄｅｒｉｖｅｄｎｅｕｒｏｔｒｏｐｈｉｃｆａｃｔｏｒｍＲＮＡｅｘｐｒｅｓｓｉｏｎｉｎｃｕｌｔｕｒｅｄａｓｔｒｏ￣ｃｙｔｅｓ[Ｊ].ＪＰｈａｒｍａｃｏｌＳｃｉꎬ２０１２ꎬ１２０(３):１７６－１８６.[１５]ＤＥＹＡＭＡＳꎬＢＡＮＧＥꎬＫＡＴＯＴꎬｅｔａｌ.ＮｅｕｒｏｔｒｏｐｈｉｃａｎｄＡｎｔｉｄｅ￣ｐｒｅｓｓａｎｔＡｃｔｉｏｎｓｏｆＢｒａｉｎ－ＤｅｒｉｖｅｄＮｅｕｒｏｔｒｏｐｈｉｃＦａｃｔｏｒＲｅｑｕｉｒｅＶａｓｃｕｌａｒＥｎｄｏｔｈｅｌｉａｌＧｒｏｗｔｈＦａｃｔｏｒ[Ｊ].ＢｉｏｌＰｓｙｃｈｉａｔｒｙꎬ２０１９ꎬ８６(２):１４３－１５２.[１６]ＨＩＮＥＳＤＪꎬＨＡＹＤＯＮＰＧ.Ａｓｔｒｏｃｙｔｉｃａｄｅｎｏｓｉｎｅ:ｆｒｏｍｓｙｎａｐｓｅｓｔｏｐｓｙｃｈｉａｔｒｉｃｄｉｓｏｒｄｅｒｓ[Ｊ].ＰｈｉｌｏｓＴｒａｎｓＲＳｏｃＬｏｎｄＢＢｉｏｌＳｃｉꎬ２０１４ꎬ３６９(１６５４):２０１３０５９４.[１７]ＭＡＪꎬＱＩＸꎬＹＡＮＧＣꎬｅｔａｌ.Ｃａｌｈｍ２ｇｏｖｅｒｎｓａｓｔｒｏｃｙｔｉｃＡＴＰｒｅｌｅａｓｉｎｇｉｎｔｈｅｄｅｖｅｌｏｐｍｅｎｔｏｆｄｅｐｒｅｓｓｉｏｎ－ｌｉｋｅｂｅｈａｖｉｏｒｓ(ｖｏｌ２２ꎬｐｇ８８３ꎬ２０１７)[Ｊ].ＭｏｌＰｓｙｃｈｉａｔｒｙꎬ２０１８ꎬ２３(４):１０９１－１０９１. [１８]ＫＩＮＯＳＨＩＴＡＭꎬＨＩＲＡＹＡＭＡＹꎬＦＵＪＩＳＨＩＴＡＫꎬｅｔａｌ.Ａｎｔｉ－Ｄｅ￣ｐｒｅｓｓａｎｔＦｌｕｏｘｅｔｉｎｅＲｅｖｅａｌｓｉｔｓＴｈｅｒａｐｅｕｔｉｃＥｆｆｅｃｔＶｉａＡｓｔｒｏｃｙｔｅｓ[Ｊ].ＥＢｉｏＭｅｄｉｃｉｎｅꎬ２０１８(３２):７２－８３.[１９]ＳＡＮＴＥＬＬＯＭꎬＶＯＬＴＥＲＲＡＡ.ＳｙｎａｐｔｉｃＭｏｄｕｌａｔｉｏｎｂｙＡｓｔｒｏｃｙｔｅｓＶｉａＣａ２＋－ＤｅｐｅｎｄｅｎｔＧｌｕｔａｍａｔｅＲｅｌｅａｓｅ[Ｊ].Ｎｅｕｒｏｓｃｉｅｎｃｅꎬ２００９ꎬ１５８(１):２５３－２５９.[２０]ＣＨＯＵＤＡＲＹＰＶꎬＭＯＬＮＡＲＭꎬＥＶＡＮＳＳＪꎬｅｔａｌ.ＡｌｔｅｒｅｄｃｏｒｔｉｃａｌｇｌｕｔａｍａｔｅｒｇｉｃａｎｄＧＡＢＡｅｒｇｉｃｓｉｇｎａｌｔｒａｎｓｍｉｓｓｉｏｎｗｉｔｈｇｌｉａｌｉｎｖｏｌｖｅ￣ｍｅｎｔｉｎｄｅｐｒｅｓｓｉｏｎ[Ｊ].ＰｒｏｃＮａｔｌＡｃａｄＳｃｉＵＳＡꎬ２００５ꎬ１０２(４３):１５６５３－１５６５８.[２１]ＳＯＮＨꎬＢＡＥＫＪＨꎬＧＯＢＳꎬｅｔａｌ.Ｇｌｕｔａｍｉｎｅｈａｓａｎｔｉｄｅｐｒｅｓｓｉｖｅｅｆｆｅｃｔｓｔｈｒｏｕｇｈｉｎｃｒｅｍｅｎｔｓｏｆｇｌｕｔａｍａｔｅａｎｄｇｌｕｔａｍｉｎｅｌｅｖｅｌｓａｎｄｇｌｕｔａｍａｔｅｒｇｉｃａｃｔｉｖｉｔｙｉｎｔｈｅｍｅｄｉａｌｐｒｅｆｒｏｎｔａｌｃｏｒｔｅｘ[Ｊ].Ｎｅｕｒｏｐ￣ｈａｒｍａｃｏｌｏｇｙꎬ２０１８(１４３):１４３－１５２.[２２]ＮＡＳＣＡＣꎬＢＩＧＩＯＢꎬＺＥＬＬＩＤꎬｅｔａｌ.ＲｏｌｅｏｆｔｈｅＡｓｔｒｏｇｌｉａｌＧｌｕｔａｍａｔｅＥｘｃｈａｎｇｅｒｘＣＴｉｎＶｅｎｔｒａｌＨｉｐｐｏｃａｍｐｕｓｉｎＲｅｓｉｌｉｅｎｃｅｔｏＳｔｒｅｓｓ[Ｊ].Ｎｅｕｒｏｎꎬ２０１７ꎬ９６(２):４０２－４１３.[２３]ＳＵＬＴＡＮＳꎬＬＩＬꎬＭＯＳＳＪꎬｅｔａｌ.ＳｙｎａｐｔｉｃＩｎｔｅｇｒａｔｉｏｎｏｆＡｄｕｌｔ－ＢｏｒｎＨｉｐｐｏｃａｍｐａｌＮｅｕｒｏｎｓＩｓＬｏｃａｌｌｙＣｏｎｔｒｏｌｌｅｄｂｙＡｓｔｒｏｃｙｔｅｓ[Ｊ].Ｎｅｕ￣ｒｏｎꎬ２０１５ꎬ８８(５):９５７－９７２.[２４]ＬＩＳＹꎬＵＮＯＹꎬＲＵＤＯＬＰＨＵꎬｅｔａｌ.Ａｓｔｒｏｃｙｔｅｓｉｎｐｒｉｍａｒｙｃｕｌｔｕｒｅｓｅｘ￣ｐｒｅｓｓｓｅｒｉｎｅｒａｃｅｍａｓｅꎬｓｙｎｔｈｅｓｉｚｅＤ－ｓｅｒｉｎｅａｎｄａｃｑｕｉｒｅＡ１ｒｅａｃｔｉｖｅａｓ￣ｔｒｏｃｙｔｅｆｅａｔｕｒｅｓ[Ｊ].ＢｉｏｃｈｅｍＰｈａｒｍａｃｏｌꎬ２０１８(５１):２４５－２５１.[２５]ＬＩＮＨꎬＪＡＣＯＢＩＡＡꎬＡＮＤＥＲＳＯＮＳＡꎬｅｔａｌ.Ｄ－ＳｅｒｉｎｅａｎｄＳｅｒｉｎｅＲａｃｅｍａｓＡｒｅＡｓｓｏｃｉａｔｅｄｗｉｔｈＰＳＤ－９５ａｎｄＧｌｕｔａｍａｔｅｒｇｉｃＳｙｎａｐｓｅＳｔａｂｉｌｉｔｙ[Ｊ].ＦｒｏｎｔＣｅｌｌＮｅｕｒｏｓｃｉꎬ２０１６(１０):３４.[２６]ＩＶＡＮＯＶＡＤꎬＭＯＴＨＥＴＪＰ.Ｔｈｅｐｌａｓｔｉｃｄ－ｓｅｒｉｎｅｓｉｇｎａｌｉｎｇｐａｔｈ￣ｗａｙ:Ｓｌｉｄｉｎｇｆｒｏｍｎｅｕｒｏｎｓｔｏｇｌｉａａｎｄｖｉｃｅ－ｖｅｒｓａ[Ｊ].ＮｅｕｒｏｓｃｉＬｅｔｔꎬ２０１９(６８９):２１－２５.[２７]ＰＥＲＥＡＧꎬＧＯＭＥＺＲꎬＭＥＤＥＲＯＳＳꎬｅｔａｌ.Ａｃｔｉｖｉｔｙ－ｄｅｐｅｎｄｅｎｔｓｗｉｔｃｈｏｆＧＡＢＡｅｒｇｉｃｉｎｈｉｂｉｔｉｏｎｉｎｔｏｇｌｕｔａｍａｔｅｒｇｉｃｅｘｃｉｔａｔｉｏｎｉｎａｓ￣ｔｒｏｃｙｔｅ－ｎｅｕｒｏｎｎｅｔｗｏｒｋｓ[Ｊ].ｅＬｉｆｅꎬ２０１６(５):ｅ２０３６２.[２８]ＣＯＶＥＬＯＡꎬＡＲＡＱＵＥＡ.Ｎｅｕｒｏｎａｌａｃｔｉｖｉｔｙｄｅｔｅｒｍｉｎｅｓｄｉｓｔｉｎｃｔｇｌｉｏ￣ｔｒａｎｓｍｉｔｔｅｒｒｅｌｅａｓｅｆｒｏｍａｓｉｎｇｌｅａｓｔｒｏｃｙｔｅ[Ｊ].ＥＬｉｆｅꎬ２０１８(７):ｅ３２２３７.[２９]ＳＨＩＭＨＳꎬＰＡＲＫＨＪꎬＷＯＯＪꎬｅｔａｌ.ＲｏｌｅｏｆａｓｔｒｏｃｙｔｉｃＧＡＢＡｅｒｇｉｃｓｙｓｔｅｍｏｎｉｎｆｌａｍｍａｔｏｒｙｃｙｔｏｋｉｎｅ－ｉｎｄｕｃｅｄａｎｘｉｅｔｙ－ｌｉｋｅｂｅｈａｖｉｏｒ[Ｊ].Ｎｅｕｒｏｐｈａｒｍａｃｏｌｏｇｙꎬ２０１９(１６０):１０７７７６.[３０]ＳＴＯＧＳＤＩＬＬＪＡꎬＲＡＭＩＲＥＺＪꎬＬＩＵＤꎬｅｔａｌ.Ａｓｔｒｏｃｙｔｉｃｎｅｕｒｏｌｉｇｉｎｓｃｏｎｔｒｏｌａｓｔｒｏｃｙｔｅｍｏｒｐｈｏｇｅｎｅｓｉｓａｎｄｓｙｎａｐｔｏｇｅｎｅｓｉｓ[Ｊ].Ｎａｔｕｒｅꎬ２０１７ꎬ５５１(７６７９):１９２－１９７.[３１]ＲＩＳＨＥＲＷＣꎬＥＲＯＧＬＵＣ.Ｔｈｒｏｍｂｏｓｐｏｎｄｉｎｓａｓｋｅｙｒｅｇｕｌａｔｏｒｓｏｆｓｙｎａｐｔｏｇｅｎｅｓｉｓｉｎｔｈｅｃｅｎｔｒａｌｎｅｒｖｏｕｓｓｙｓｔｅｍ[Ｊ].ＭａｔｒｉｘＢｉｏｌꎬ２０１２ꎬ３１(３):１７０－１７７.[３２]ＬＩＮꎬＬＩＵＲＪꎬＤＷＹＥＲＪＭꎬｅｔａｌ.ＧｌｕｔａｍａｔｅＮ－ｍｅｔｈｙｌ－Ｄ－ａｓｐａｒｔａｔｅｒｅｃｅｐｔｏｒａｎｔａｇｏｎｉｓｔｓｒａｐｉｄｌｙｒｅｖｅｒｓｅｂｅｈａｖｉｏｒａｌａｎｄｓｙｎａｐｔｉｃｄｅｆｉｃｉｔｓｃａｕｓｅｄｂｙｃｈｒｏｎｉｃｓｔｒｅｓｓｅｘｐｏｓｕｒｅ[Ｊ].ＢｉｏｌＰｓｙｃｈｉ￣ａｔｒｙꎬ２０１１ꎬ６９(８):７５４－７６１.(下转第１６８页)１９ꎬ２０－Ｚ－ｖａｌｌｅｓａｍｉｎｅａｎｄ１９ꎬ２０－Ｅ－ｖａｌｌｅｓａｍｉｎｅｆｒｏｍＡｌｓｔｏｎｉａｓｃｈｏｌａｒｉｓ[Ｊ].Ｈｅｔｅｒｏｃｙｃｌｅｓꎬ１９８７(２６):４１３－４１９.[４４]ＺＨＵＷＭꎬＨＥＨＰꎬＦＡＮＬＭꎬｅｔａｌ.Ｃｏｍｐｏｎｅｎｔｓｏｆｓｔｅｍｂａｒｋｓｏｆ￣ＷｉｎｃｈｉａｃａｌｏｐｈｙｌｌａＡ.ＤＣ.ａｎｄｔｈｅｉｒｂｒｏｎｃｈｏｄｉｌａｔｏｒａｃｔｉｖｉｔｉｅｓ[Ｊ].ＪＩｎｔｅｇｒａｔｉｖｅＰｌａｎｔＢｉｏｌｏｇｙꎬ２００５ꎬ４７(７):８９２－８９６.[４５]ＰＡＴＲＡＡꎬＧＨＯＳＨＧꎬＳＥＮＧＵＰＴＡＰＫꎬｅｔａｌ.Ｃａｒｂｏｎ－１３ＮＭＲｓｐｅｃｔｒａｌｓｔｕｄｉｅｓｏｎｃｈａｌｃｏｎｅｓａｎｄａｃｅｔｏｐｈｅｎｏｎｅｓ[Ｊ].ＭａｇｎＲｅｓｏｎＣｈｅｍꎬ１９８７ꎬ２５(８):７３４－７３６.[４６]ＳＣＯＴＴＫＮ.Ｃａｒｂｏｎ－１３ｎｕｃｌｅａｒｍａｇｎｅｔｉｃｒｅｓｏｎａｎｃｅｏｆｂｉｏｌｏｇｉｃａｌｌｙｉｍｐｏｒｔａｎｔａｒｏｍａｔｉｃａｃｉｄｓ.Ｉ.Ｃｈｅｍｉｃａｌｓｈｉｆｔｓｏｆｂｅｎｚｏｉｃａｃｉｄａｎｄｄｅ￣ｒｉｖａｔｉｖｅｓ[Ｊ].ＪＡｍＣｈｅｍＳｏｃꎬ１９７２ꎬ９４(２４):８５６４－８５６８.[４７]ＯＨＡＳＨＩＫꎬＷＡＴＡＮＡＢＥＨꎬＯＫＵＭＵＲＡＹꎬｅｔａｌ.Ｉｎｄｏｎｅｓｉａｎｍｅ￣ｄｉｃｉｎａｌｐｌａｎｔｓ.ＸＩＩ.Ｆｏｕｒｉｓｏｍａｒｉｃｌｉｇｎａｎ－ｇｌｕｃｏｓｉｄｅｓｆｒｏｍｔｈｅｂａｒｋｏｆＡｅｇｌｅｍａｒｍｅｌｏｓ(Ｒｕｔａｃｅａｅ)[Ｊ].ＣｈｅｍＰｈａｒｍＢｕｌｌꎬ１９９４(４２):１９２４－１９２６.[４８]ＹＡＭＡＵＣＨＩＴꎬＡＢＥＦꎬＣＨＥＮＲＦꎬｅｔａｌ.ＡｌｋａｌｏｉｄｓｆｒｏｍｔｈｅｌｅａｖｅｓｏｆＡｌｓｔｏｎｉａｓｃｈｏｌａｒｉｓｉｎＴａｉｗａｎꎬＴｈａｉｌａｎｄꎬＩｎｄｏｎｅｓｉａａｎｄｔｈｅＰｈｉｌｉｐ￣ｐｉｎｅｓ[Ｊ].Ｐｈｙｔｏｃｈｅｍｉｓｔｒｙꎬ１９９０ꎬ２９(１１):３５４７－３５５２.[４９]ＡＴＴＡ－ＵＲ－ＲＡＨＭＡＮꎬＡＬＶＩＫＡ.ＩｎｄｏｌｅａｌｋａｌｏｉｄｓｆｒｏｍＡｌｓｔｏｎｉａｓｃｈｏｌａｒｉｓ[Ｊ].Ｐｈｙｔｏｃｈｅｍｉｓｔｒｙꎬ１９８７ꎬ２６(７):２１３９－２１４２.[５０]ＡＢＥＦꎬＣＨＥＮＲＦꎬＹＡＭＡＵＣＨＩＴꎬｅｔａｌ.Ａｌｓｃｈｏｍｉｎｅａｎｄｉｓｏａｌｓｃｈｏ￣ｍｉｎｅꎬｎｅｗａｌｋａｌｏｉｄｓｆｒｏｍｔｈｅｌｅａｖｅｓｏｆＡｌｓｔｏｎｉａｓｃｈｏｌａｒｉｓ[Ｊ].ＣｈｅｍＰｈａｒｍＢｕｌｌꎬ１９８９ꎬ３７(４):８８７－８９０.[５１]ＡＣＨＥＮＢＡＣＨＨꎬＢＥＮＩＲＳＣＨＫＥＭꎬＴＯＲＲＥＮＥＧＲＡＲ.ＡｌｋａｌｏｉｄｓａｎｄｏｔｈｅｒｃｏｍｐｏｕｎｄｓｆｒｏｍｓｅｅｄｓｏｆＴａｂｅｒｎａｅｍｏｎｔａｎａｃｙｍｏｓａ[Ｊ].Ｐｈｙｔｏｃｈｅｍｉｓｔｒｙꎬ１９９７ꎬ４５(２):３２５－３３５.[５２]ＴＨＥＲＥＳＥＲＡＶＡＯꎬＲＩＣＨＡＲＤＢꎬＺＥＣＨＥＳＭꎬｅｔａｌ.Ｔｈｅｃｏｎｆｉｇｕ￣ｒａｔｉｏｎｏｆｖｅｎｏｔｅｒｐｉｎｅ[Ｊ].ＴｅｔｒａｈｅｄｒｏｎＬｅｔｔｅｒｓꎬ１９８５ꎬ２６(７):８３７－８３８.[５３]张慧燕ꎬ阎文玫ꎬ陈德昌.吡啶单萜烯ｉｓｏｃａｎｔｌｅｙｉｎｅ的结构测定[Ｊ].药学学报ꎬ１９９２(２):１１３－１１６[５４]唐元清ꎬ冯孝章ꎬ黄量.吴茱萸化学成分的研究[Ｊ].药学学报ꎬ１９９６ꎬ３１(２):１５１－１５５.[５５]ＡＲＩＳＡＷＡＭꎬＨＯＲＩＵＣＨＩＴꎬＨＡＹＡＳＨＩＴꎬｅｔａｌ.Ｓｔｕｄｉｅｓｏｎｃｏｎｓｔｉｔ￣ｕｅｎｔｓｏｆＥｖｏｄｉａｒｕｔａｅｃａｒｐａ(Ｒｕｔａｃｅａｅ)[Ｊ].ＣｈｅｍＰｈａｒｍＢｕｌｌꎬ１９９３ꎬ４１(８):１４７２－１４７４.[５６]ＴＡＮＧＹＱꎬＦＥＮＧＸＺꎬＨＵＡＮＧＬ.ＱｕｉｎｏｌｏｎｅａｌｋａｌｏｉｄｓｆｒｏｍＥｖｏｄｉａｒｕｔａｅｃａｒｐａ[Ｊ].Ｐｈｙｔｏｃｈｅｍｉｓｔｒｙꎬ１９９６ꎬ４３(３):７１９－７２２.[５７]ＣＡＩＸＨꎬＢＡＯＭＦꎬＺＨＡＮＧＹꎬｅｔａｌ.ＡｎｅｗｔｙｐｅｏｆｍｏｎｏｔｅｒｐｅｎｏｉｄｉｎｄｏｌｅａｌｋａｌｏｉｄｐｒｅｃｕｒｓｏｒｆｒｏｍＡｌｓｔｏｎｉａｒｏｓｔｒａｔａ[Ｊ].ＯｒｇａｎｉｃＬｅｔｔｅｒｓꎬ２０１１ꎬ１３(１４):３５６８－３５７１.[５８]ＢＡＯＭＦꎬＺＥＮＧＣＸꎬＱＵＹꎬｅｔａｌ.ＭｏｎｏｔｅｒｐｅｎｏｉｄｉｎｄｏｌｅａｌｋａｌｏｉｄｓｆｒｏｍＡｌｓｔｏｎｉａｒｏｓｔｒａｔａ[Ｊ].ＮａｔＰｒｏＢｉｏꎬ２０１２ꎬ２(３):１２１－１２５. [５９]ＺＨＯＮＧＸＨꎬＢＡＯＭＦꎬＺＥＮＧＣＸꎬｅｔａｌ.ＰｏｌｙｃｙｃｌｉｃｍｏｎｏｔｅｒｐｅｎｏｉｄｉｎｄｏｌｅａｌｋａｌｏｉｄｓｆｒｏｍＡｌｓｔｏｎｉａｒｏｓｔｒａｔａａｎｄｔｈｅｉｒｒｅｔｉｃｕｌａｔｅｄｅｒｉｖａｔｉｏｎ[Ｊ].ＰｈｙＬｅｔｔꎬ２０１７(２０):７７－８３.[６０]ＹＵＡＮＹＸꎬＧＵＯＦꎬＨＥＨＰꎬｅｔａｌ.ＴｗｏｎｅｗｍｏｎｏｔｅｒｐｅｎｏｉｄｉｎｄｏｌｅａｌｋａｌｏｉｄｓｆｒｏｍＡｌｓｔｏｎｉａｒｏｓｔｒａｔａ[Ｊ].ＮａｔＰｒｏＬｅｔｔꎬ２０１８ꎬ３２(７):８４４－８４８.[６１]ＬＩＭＪＬꎬＳＩＭＫＳꎬＹＯＮＧＫＴꎬｅｔａｌ.ＢｉｏｌｏｇｉｃａｌｌｙａｃｔｉｖｅｖａｌｌｅｓａｍｉｎｅꎬｓｔｒｙｃｈｎａｎꎬａｎｄｒｈａｚｉｎｉｌａｍａｌｋａｌｏｉｄｓｆｒｏｍＡｌｓｔｏｎｉａＰｎｅｕｍａｔｏｐｈｏｒｉｎｅꎬａｎｏｒ－ｓｅｃｏｖａｌｌｅｓａｍｉｎｅｗｉｔｈｕｎｕｓｕａｌｉｎｃｏｒｐｏｒａｔｉｏｎｏｆａ３－ｅｔｈｙｌｐｙｒｉｄｉｎｅｍｏｉｅｔｙ[Ｊ].Ｐｈｙｔｏｃｈｅｍｉｓｔｒｙꎬ２０１５(１１７):３１７－３２４.[６２]ＡＬＬＥＹＭＣꎬＳＣＵＤＩＥＲＯＤＡꎬＭＯＮＫＳＡꎬｅｔａｌ.ＦｅａｓｉｂｉｌｉｔｙｏｆＤｒｕｇＳｃｒｅｅｎｉｎｇｗｉｔｈＰａｎｅｌｓｏｆＨｕｍａｎＴｕｍｏｒＣｅｌｌＬｉｎｅｓＵｓｉｎｇａＭｉｃｒｏ￣ｃｕｌｔｕｒｅＴｅｔｒａｚｏｌｉｕｍＡｓｓａｙ[Ｊ].ＣａｎｃｅｒＲｅｓꎬ１９８８ꎬ４８(３):５８９－６０１. [６３]ＥＬＬＭＡＮＧＬꎬＣＯＵＲＴＮＥＹＫＤꎬＡＮＤＲＥＳＶꎬｅｔａｌ.ＡＮｅｗａｎｄＲａｐｉｄＣｏｌｏｒｉｍｅｔｒｉｃＤｅｔｅｒｍｉｎａｔｉｏｎｏｆＡｃｅｔｙｌｃｈｏｌｉｎｅｓｔｅｒａｓｅＡｃｔｉｖｉｔｙ[Ｊ].ＢｉｏｃｈｅｍＰｈａｒｍａｃｏｌꎬ１９６１(７):８８－９５.(上接第１５９页)[３３]ＢＡＲＫＥＲＡＪꎬＵＬＬＩＡＮＥＭ.ＡｓｔｒｏｃｙｔｅｓａｎｄＳｙｎａｐｔｉｃＰｌａｓｔｉｃｉｔｙ[Ｊ].Ｎｅｕｒｏｓｃｉｅｎｔｉｓｔꎬ２０１０ꎬ１６(１):４０－５０. [３４]ＥＲＮＳＴＣꎬＮＡＧＹＣꎬＫＩＭＳꎬｅｔａｌ.ＤｙｓｆｕｎｃｔｉｏｎｏｆＡｓｔｒｏｃｙｔｅＣｏｎｎｅｘｉｎｓ３０ａｎｄ４３ｉｎＤｏｒｓａｌＬａｔｅｒａｌＰｒｅｆｒｏｎｔａｌＣｏｒｔｅｘｏｆＳｕｉｃｉｄｅＣｏｍｐｌｅｔｅｒｓ[Ｊ].ＢｉｏｌＰｓｙｃｈｉａｔｒｙꎬ２０１１ꎬ７０(４):３１２－３１９.[３５]ＳＵＮＪＤꎬＬＩＵＹꎬＹＵＡＮＹＨꎬｅｔａｌ.ＧａｐＪｕｎｃｔｉｏｎＤｙｓ￣ｆｕｎｃｔｉｏｎｉｎｔｈｅＰｒｅｆｒｏｎｔａｌＣｏｒｔｅｘＩｎｄｕｃｅｓＤｅｐｒｅｓｓｉｖｅ－ＬｉｋｅＢｅｈａｖｉｏｒｓｉｎＲａｔｓ[Ｊ].Ｎｅｕｒｏｐｓｙｃｈｏｐｈａｒｍａｃｏｌｏｇｙꎬ２０１２ꎬ３７(５):１３０５－１３２０.[３６]ＭＯＲＩＯＫＡＮꎬＳＵＥＫＡＭＡＫꎬＺＨＡＮＧＦＦꎬｅｔａｌ.Ａｍｉｔｒｉｐ￣ｔｙｌｉｎｅｕｐ－ｒｅｇｕｌａｔｅｓｃｏｎｎｅｘｉｎ４３－ｇａｐｊｕｎｃｔｉｏｎｉｎｒａｔｃｕｌ￣ｔｕｒｅｄｃｏｒｔｉｃａｌａｓｔｒｏｃｙｔｅｓｖｉａａｃｔｉｖａｔｉｏｎｏｆｔｈｅｐ３８ａｎｄｃ－Ｆｏｓ/ＡＰ－１ｓｉｇｎａｌｌｉｎｇｐａｔｈｗａｙ[Ｊ].ＢｒＪＰｈａｒｍａｃｏｌꎬ２０１４ꎬ１７１(１１):２８５４－２８６７.[３７]ＫＯＤＡＭＡＭꎬＦＵＪＩＯＫＡＴꎬＤＵＭＡＮＲＳ.Ｃｈｒｏｎｉｃｏｌａｎｚａ￣ｐｉｎｅｏｒｆｌｕｏｘｅｔｉｎｅａｄｍｉｎｉｓｔｒａｔｉｏｎｉｎｃｒｅａｓｅｓｃｅｌｌｐｒｏｌｉｆｅｒａｔｉｏｎｉｎｈｉｐｐｏｃａｍｐｕｓａｎｄｐｒｅｆｒｏｎｔａｌｃｏｒｔｅｘｏｆａｄｕｌｔｒａｔ[Ｊ].ＢｉｏｌＰｓｙｃｈｉａｔｒｙꎬ２００４ꎬ５６(８):５７０－５８０. [３８]ＬＩＡＮＧＭꎬＺＨＯＮＧＨꎬＲＯＮＧＪꎬｅｔａｌ.ＰｏｓｔｎａｔａｌＬｉｐｏｐｏ￣ｌｙｓａｃｃｈａｒｉｄｅＥｘｐｏｓｕｒｅＩｍｐａｉｒｓＡｄｕｌｔＮｅｕｒｏｇｅｎｅｓｉｓａｎｄＣａｕｓｅｓＤｅｐｒｅｓｓｉｏｎ－ｌｉｋｅＢｅｈａｖｉｏｒｓＴｈｒｏｕｇｈＡｓｔｒｏｃｙｔｅｓＡｃｔｉｖａｔｉｏｎＴｒｉｇｇｅｒｉｎｇＧＡＢＡＡＲｅｃｅｐｔｏｒＤｏｗｎｒｅｇｕｌａｔｉｏｎ[Ｊ].Ｎｅｕｒｏｓｃｉｅｎｃｅꎬ２０１９(４２２):２１－３１.[３９]ＱＵＥＳＳＥＶＥＵＲＧꎬＤＡＶＩＤＤＪꎬＧＡＩＬＬＡＲＤＭＣꎬｅｔａｌ.ＢＤＮＦｏｖｅｒｅｘｐｒｅｓｓｉｏｎｉｎｍｏｕｓｅｈｉｐｐｏｃａｍｐａｌａｓｔｒｏｃｙｔｅｓｐｒｏｍｏｔｅｓｌｏｃａｌｎｅｕｒｏｇｅｎｅｓｉｓａｎｄｅｌｉｃｉｔｓａｎｘｉｏｌｙｔｉｃ－ｌｉｋｅａｃｔｉｖｉｔｉｅｓ[Ｊ].ＴｒａｎｓｌＰｓｙｃｈｉａｔｒｙꎬ２０１３(３):ｅ２５３.。

Draft:Deep Learning in Neural Networks:An OverviewTechnical Report IDSIA-03-14/arXiv:1404.7828(v1.5)[cs.NE]J¨u rgen SchmidhuberThe Swiss AI Lab IDSIAIstituto Dalle Molle di Studi sull’Intelligenza ArtificialeUniversity of Lugano&SUPSIGalleria2,6928Manno-LuganoSwitzerland15May2014AbstractIn recent years,deep artificial neural networks(including recurrent ones)have won numerous con-tests in pattern recognition and machine learning.This historical survey compactly summarises relevantwork,much of it from the previous millennium.Shallow and deep learners are distinguished by thedepth of their credit assignment paths,which are chains of possibly learnable,causal links between ac-tions and effects.I review deep supervised learning(also recapitulating the history of backpropagation),unsupervised learning,reinforcement learning&evolutionary computation,and indirect search for shortprograms encoding deep and large networks.PDF of earlier draft(v1):http://www.idsia.ch/∼juergen/DeepLearning30April2014.pdfLATEX source:http://www.idsia.ch/∼juergen/DeepLearning30April2014.texComplete BIBTEXfile:http://www.idsia.ch/∼juergen/bib.bibPrefaceThis is the draft of an invited Deep Learning(DL)overview.One of its goals is to assign credit to those who contributed to the present state of the art.I acknowledge the limitations of attempting to achieve this goal.The DL research community itself may be viewed as a continually evolving,deep network of scientists who have influenced each other in complex ways.Starting from recent DL results,I tried to trace back the origins of relevant ideas through the past half century and beyond,sometimes using“local search”to follow citations of citations backwards in time.Since not all DL publications properly acknowledge earlier relevant work,additional global search strategies were employed,aided by consulting numerous neural network experts.As a result,the present draft mostly consists of references(about800entries so far).Nevertheless,through an expert selection bias I may have missed important work.A related bias was surely introduced by my special familiarity with the work of my own DL research group in the past quarter-century.For these reasons,the present draft should be viewed as merely a snapshot of an ongoing credit assignment process.To help improve it,please do not hesitate to send corrections and suggestions to juergen@idsia.ch.Contents1Introduction to Deep Learning(DL)in Neural Networks(NNs)3 2Event-Oriented Notation for Activation Spreading in FNNs/RNNs3 3Depth of Credit Assignment Paths(CAPs)and of Problems4 4Recurring Themes of Deep Learning54.1Dynamic Programming(DP)for DL (5)4.2Unsupervised Learning(UL)Facilitating Supervised Learning(SL)and RL (6)4.3Occam’s Razor:Compression and Minimum Description Length(MDL) (6)4.4Learning Hierarchical Representations Through Deep SL,UL,RL (6)4.5Fast Graphics Processing Units(GPUs)for DL in NNs (6)5Supervised NNs,Some Helped by Unsupervised NNs75.11940s and Earlier (7)5.2Around1960:More Neurobiological Inspiration for DL (7)5.31965:Deep Networks Based on the Group Method of Data Handling(GMDH) (8)5.41979:Convolution+Weight Replication+Winner-Take-All(WTA) (8)5.51960-1981and Beyond:Development of Backpropagation(BP)for NNs (8)5.5.1BP for Weight-Sharing Feedforward NNs(FNNs)and Recurrent NNs(RNNs)..95.6Late1980s-2000:Numerous Improvements of NNs (9)5.6.1Ideas for Dealing with Long Time Lags and Deep CAPs (10)5.6.2Better BP Through Advanced Gradient Descent (10)5.6.3Discovering Low-Complexity,Problem-Solving NNs (11)5.6.4Potential Benefits of UL for SL (11)5.71987:UL Through Autoencoder(AE)Hierarchies (12)5.81989:BP for Convolutional NNs(CNNs) (13)5.91991:Fundamental Deep Learning Problem of Gradient Descent (13)5.101991:UL-Based History Compression Through a Deep Hierarchy of RNNs (14)5.111992:Max-Pooling(MP):Towards MPCNNs (14)5.121994:Contest-Winning Not So Deep NNs (15)5.131995:Supervised Recurrent Very Deep Learner(LSTM RNN) (15)5.142003:More Contest-Winning/Record-Setting,Often Not So Deep NNs (16)5.152006/7:Deep Belief Networks(DBNs)&AE Stacks Fine-Tuned by BP (17)5.162006/7:Improved CNNs/GPU-CNNs/BP-Trained MPCNNs (17)5.172009:First Official Competitions Won by RNNs,and with MPCNNs (18)5.182010:Plain Backprop(+Distortions)on GPU Yields Excellent Results (18)5.192011:MPCNNs on GPU Achieve Superhuman Vision Performance (18)5.202011:Hessian-Free Optimization for RNNs (19)5.212012:First Contests Won on ImageNet&Object Detection&Segmentation (19)5.222013-:More Contests and Benchmark Records (20)5.22.1Currently Successful Supervised Techniques:LSTM RNNs/GPU-MPCNNs (21)5.23Recent Tricks for Improving SL Deep NNs(Compare Sec.5.6.2,5.6.3) (21)5.24Consequences for Neuroscience (22)5.25DL with Spiking Neurons? (22)6DL in FNNs and RNNs for Reinforcement Learning(RL)236.1RL Through NN World Models Yields RNNs With Deep CAPs (23)6.2Deep FNNs for Traditional RL and Markov Decision Processes(MDPs) (24)6.3Deep RL RNNs for Partially Observable MDPs(POMDPs) (24)6.4RL Facilitated by Deep UL in FNNs and RNNs (25)6.5Deep Hierarchical RL(HRL)and Subgoal Learning with FNNs and RNNs (25)6.6Deep RL by Direct NN Search/Policy Gradients/Evolution (25)6.7Deep RL by Indirect Policy Search/Compressed NN Search (26)6.8Universal RL (27)7Conclusion271Introduction to Deep Learning(DL)in Neural Networks(NNs) Which modifiable components of a learning system are responsible for its success or failure?What changes to them improve performance?This has been called the fundamental credit assignment problem(Minsky, 1963).There are general credit assignment methods for universal problem solvers that are time-optimal in various theoretical senses(Sec.6.8).The present survey,however,will focus on the narrower,but now commercially important,subfield of Deep Learning(DL)in Artificial Neural Networks(NNs).We are interested in accurate credit assignment across possibly many,often nonlinear,computational stages of NNs.Shallow NN-like models have been around for many decades if not centuries(Sec.5.1).Models with several successive nonlinear layers of neurons date back at least to the1960s(Sec.5.3)and1970s(Sec.5.5). An efficient gradient descent method for teacher-based Supervised Learning(SL)in discrete,differentiable networks of arbitrary depth called backpropagation(BP)was developed in the1960s and1970s,and ap-plied to NNs in1981(Sec.5.5).BP-based training of deep NNs with many layers,however,had been found to be difficult in practice by the late1980s(Sec.5.6),and had become an explicit research subject by the early1990s(Sec.5.9).DL became practically feasible to some extent through the help of Unsupervised Learning(UL)(e.g.,Sec.5.10,5.15).The1990s and2000s also saw many improvements of purely super-vised DL(Sec.5).In the new millennium,deep NNs havefinally attracted wide-spread attention,mainly by outperforming alternative machine learning methods such as kernel machines(Vapnik,1995;Sch¨o lkopf et al.,1998)in numerous important applications.In fact,supervised deep NNs have won numerous of-ficial international pattern recognition competitions(e.g.,Sec.5.17,5.19,5.21,5.22),achieving thefirst superhuman visual pattern recognition results in limited domains(Sec.5.19).Deep NNs also have become relevant for the more generalfield of Reinforcement Learning(RL)where there is no supervising teacher (Sec.6).Both feedforward(acyclic)NNs(FNNs)and recurrent(cyclic)NNs(RNNs)have won contests(Sec.5.12,5.14,5.17,5.19,5.21,5.22).In a sense,RNNs are the deepest of all NNs(Sec.3)—they are general computers more powerful than FNNs,and can in principle create and process memories of ar-bitrary sequences of input patterns(e.g.,Siegelmann and Sontag,1991;Schmidhuber,1990a).Unlike traditional methods for automatic sequential program synthesis(e.g.,Waldinger and Lee,1969;Balzer, 1985;Soloway,1986;Deville and Lau,1994),RNNs can learn programs that mix sequential and parallel information processing in a natural and efficient way,exploiting the massive parallelism viewed as crucial for sustaining the rapid decline of computation cost observed over the past75years.The rest of this paper is structured as follows.Sec.2introduces a compact,event-oriented notation that is simple yet general enough to accommodate both FNNs and RNNs.Sec.3introduces the concept of Credit Assignment Paths(CAPs)to measure whether learning in a given NN application is of the deep or shallow type.Sec.4lists recurring themes of DL in SL,UL,and RL.Sec.5focuses on SL and UL,and on how UL can facilitate SL,although pure SL has become dominant in recent competitions(Sec.5.17-5.22). Sec.5is arranged in a historical timeline format with subsections on important inspirations and technical contributions.Sec.6on deep RL discusses traditional Dynamic Programming(DP)-based RL combined with gradient-based search techniques for SL or UL in deep NNs,as well as general methods for direct and indirect search in the weight space of deep FNNs and RNNs,including successful policy gradient and evolutionary methods.2Event-Oriented Notation for Activation Spreading in FNNs/RNNs Throughout this paper,let i,j,k,t,p,q,r denote positive integer variables assuming ranges implicit in the given contexts.Let n,m,T denote positive integer constants.An NN’s topology may change over time(e.g.,Fahlman,1991;Ring,1991;Weng et al.,1992;Fritzke, 1994).At any given moment,it can be described as afinite subset of units(or nodes or neurons)N= {u1,u2,...,}and afinite set H⊆N×N of directed edges or connections between nodes.FNNs are acyclic graphs,RNNs cyclic.Thefirst(input)layer is the set of input units,a subset of N.In FNNs,the k-th layer(k>1)is the set of all nodes u∈N such that there is an edge path of length k−1(but no longer path)between some input unit and u.There may be shortcut connections between distant layers.The NN’s behavior or program is determined by a set of real-valued,possibly modifiable,parameters or weights w i(i=1,...,n).We now focus on a singlefinite episode or epoch of information processing and activation spreading,without learning through weight changes.The following slightly unconventional notation is designed to compactly describe what is happening during the runtime of the system.During an episode,there is a partially causal sequence x t(t=1,...,T)of real values that I call events.Each x t is either an input set by the environment,or the activation of a unit that may directly depend on other x k(k<t)through a current NN topology-dependent set in t of indices k representing incoming causal connections or links.Let the function v encode topology information and map such event index pairs(k,t)to weight indices.For example,in the non-input case we may have x t=f t(net t)with real-valued net t= k∈in t x k w v(k,t)(additive case)or net t= k∈in t x k w v(k,t)(multiplicative case), where f t is a typically nonlinear real-valued activation function such as tanh.In many recent competition-winning NNs(Sec.5.19,5.21,5.22)there also are events of the type x t=max k∈int (x k);some networktypes may also use complex polynomial activation functions(Sec.5.3).x t may directly affect certain x k(k>t)through outgoing connections or links represented through a current set out t of indices k with t∈in k.Some non-input events are called output events.Note that many of the x t may refer to different,time-varying activations of the same unit in sequence-processing RNNs(e.g.,Williams,1989,“unfolding in time”),or also in FNNs sequentially exposed to time-varying input patterns of a large training set encoded as input events.During an episode,the same weight may get reused over and over again in topology-dependent ways,e.g.,in RNNs,or in convolutional NNs(Sec.5.4,5.8).I call this weight sharing across space and/or time.Weight sharing may greatly reduce the NN’s descriptive complexity,which is the number of bits of information required to describe the NN (Sec.4.3).In Supervised Learning(SL),certain NN output events x t may be associated with teacher-given,real-valued labels or targets d t yielding errors e t,e.g.,e t=1/2(x t−d t)2.A typical goal of supervised NN training is tofind weights that yield episodes with small total error E,the sum of all such e t.The hope is that the NN will generalize well in later episodes,causing only small errors on previously unseen sequences of input events.Many alternative error functions for SL and UL are possible.SL assumes that input events are independent of earlier output events(which may affect the environ-ment through actions causing subsequent perceptions).This assumption does not hold in the broaderfields of Sequential Decision Making and Reinforcement Learning(RL)(Kaelbling et al.,1996;Sutton and Barto, 1998;Hutter,2005)(Sec.6).In RL,some of the input events may encode real-valued reward signals given by the environment,and a typical goal is tofind weights that yield episodes with a high sum of reward signals,through sequences of appropriate output actions.Sec.5.5will use the notation above to compactly describe a central algorithm of DL,namely,back-propagation(BP)for supervised weight-sharing FNNs and RNNs.(FNNs may be viewed as RNNs with certainfixed zero weights.)Sec.6will address the more general RL case.3Depth of Credit Assignment Paths(CAPs)and of ProblemsTo measure whether credit assignment in a given NN application is of the deep or shallow type,I introduce the concept of Credit Assignment Paths or CAPs,which are chains of possibly causal links between events.Let usfirst focus on SL.Consider two events x p and x q(1≤p<q≤T).Depending on the appli-cation,they may have a Potential Direct Causal Connection(PDCC)expressed by the Boolean predicate pdcc(p,q),which is true if and only if p∈in q.Then the2-element list(p,q)is defined to be a CAP from p to q(a minimal one).A learning algorithm may be allowed to change w v(p,q)to improve performance in future episodes.More general,possibly indirect,Potential Causal Connections(PCC)are expressed by the recursively defined Boolean predicate pcc(p,q),which in the SL case is true only if pdcc(p,q),or if pcc(p,k)for some k and pdcc(k,q).In the latter case,appending q to any CAP from p to k yields a CAP from p to q(this is a recursive definition,too).The set of such CAPs may be large but isfinite.Note that the same weight may affect many different PDCCs between successive events listed by a given CAP,e.g.,in the case of RNNs, or weight-sharing FNNs.Suppose a CAP has the form(...,k,t,...,q),where k and t(possibly t=q)are thefirst successive elements with modifiable w v(k,t).Then the length of the suffix list(t,...,q)is called the CAP’s depth (which is0if there are no modifiable links at all).This depth limits how far backwards credit assignment can move down the causal chain tofind a modifiable weight.1Suppose an episode and its event sequence x1,...,x T satisfy a computable criterion used to decide whether a given problem has been solved(e.g.,total error E below some threshold).Then the set of used weights is called a solution to the problem,and the depth of the deepest CAP within the sequence is called the solution’s depth.There may be other solutions(yielding different event sequences)with different depths.Given somefixed NN topology,the smallest depth of any solution is called the problem’s depth.Sometimes we also speak of the depth of an architecture:SL FNNs withfixed topology imply a problem-independent maximal problem depth bounded by the number of non-input layers.Certain SL RNNs withfixed weights for all connections except those to output units(Jaeger,2001;Maass et al.,2002; Jaeger,2004;Schrauwen et al.,2007)have a maximal problem depth of1,because only thefinal links in the corresponding CAPs are modifiable.In general,however,RNNs may learn to solve problems of potentially unlimited depth.Note that the definitions above are solely based on the depths of causal chains,and agnostic of the temporal distance between events.For example,shallow FNNs perceiving large“time windows”of in-put events may correctly classify long input sequences through appropriate output events,and thus solve shallow problems involving long time lags between relevant events.At which problem depth does Shallow Learning end,and Deep Learning begin?Discussions with DL experts have not yet yielded a conclusive response to this question.Instead of committing myself to a precise answer,let me just define for the purposes of this overview:problems of depth>10require Very Deep Learning.The difficulty of a problem may have little to do with its depth.Some NNs can quickly learn to solve certain deep problems,e.g.,through random weight guessing(Sec.5.9)or other types of direct search (Sec.6.6)or indirect search(Sec.6.7)in weight space,or through training an NNfirst on shallow problems whose solutions may then generalize to deep problems,or through collapsing sequences of(non)linear operations into a single(non)linear operation—but see an analysis of non-trivial aspects of deep linear networks(Baldi and Hornik,1994,Section B).In general,however,finding an NN that precisely models a given training set is an NP-complete problem(Judd,1990;Blum and Rivest,1992),also in the case of deep NNs(S´ıma,1994;de Souto et al.,1999;Windisch,2005);compare a survey of negative results(S´ıma, 2002,Section1).Above we have focused on SL.In the more general case of RL in unknown environments,pcc(p,q) is also true if x p is an output event and x q any later input event—any action may affect the environment and thus any later perception.(In the real world,the environment may even influence non-input events computed on a physical hardware entangled with the entire universe,but this is ignored here.)It is possible to model and replace such unmodifiable environmental PCCs through a part of the NN that has already learned to predict(through some of its units)input events(including reward signals)from former input events and actions(Sec.6.1).Its weights are frozen,but can help to assign credit to other,still modifiable weights used to compute actions(Sec.6.1).This approach may lead to very deep CAPs though.Some DL research is about automatically rephrasing problems such that their depth is reduced(Sec.4). In particular,sometimes UL is used to make SL problems less deep,e.g.,Sec.5.10.Often Dynamic Programming(Sec.4.1)is used to facilitate certain traditional RL problems,e.g.,Sec.6.2.Sec.5focuses on CAPs for SL,Sec.6on the more complex case of RL.4Recurring Themes of Deep Learning4.1Dynamic Programming(DP)for DLOne recurring theme of DL is Dynamic Programming(DP)(Bellman,1957),which can help to facili-tate credit assignment under certain assumptions.For example,in SL NNs,backpropagation itself can 1An alternative would be to count only modifiable links when measuring depth.In many typical NN applications this would not make a difference,but in some it would,e.g.,Sec.6.1.be viewed as a DP-derived method(Sec.5.5).In traditional RL based on strong Markovian assumptions, DP-derived methods can help to greatly reduce problem depth(Sec.6.2).DP algorithms are also essen-tial for systems that combine concepts of NNs and graphical models,such as Hidden Markov Models (HMMs)(Stratonovich,1960;Baum and Petrie,1966)and Expectation Maximization(EM)(Dempster et al.,1977),e.g.,(Bottou,1991;Bengio,1991;Bourlard and Morgan,1994;Baldi and Chauvin,1996; Jordan and Sejnowski,2001;Bishop,2006;Poon and Domingos,2011;Dahl et al.,2012;Hinton et al., 2012a).4.2Unsupervised Learning(UL)Facilitating Supervised Learning(SL)and RL Another recurring theme is how UL can facilitate both SL(Sec.5)and RL(Sec.6).UL(Sec.5.6.4) is normally used to encode raw incoming data such as video or speech streams in a form that is more convenient for subsequent goal-directed learning.In particular,codes that describe the original data in a less redundant or more compact way can be fed into SL(Sec.5.10,5.15)or RL machines(Sec.6.4),whose search spaces may thus become smaller(and whose CAPs shallower)than those necessary for dealing with the raw data.UL is closely connected to the topics of regularization and compression(Sec.4.3,5.6.3). 4.3Occam’s Razor:Compression and Minimum Description Length(MDL) Occam’s razor favors simple solutions over complex ones.Given some programming language,the prin-ciple of Minimum Description Length(MDL)can be used to measure the complexity of a solution candi-date by the length of the shortest program that computes it(e.g.,Solomonoff,1964;Kolmogorov,1965b; Chaitin,1966;Wallace and Boulton,1968;Levin,1973a;Rissanen,1986;Blumer et al.,1987;Li and Vit´a nyi,1997;Gr¨u nwald et al.,2005).Some methods explicitly take into account program runtime(Al-lender,1992;Watanabe,1992;Schmidhuber,2002,1995);many consider only programs with constant runtime,written in non-universal programming languages(e.g.,Rissanen,1986;Hinton and van Camp, 1993).In the NN case,the MDL principle suggests that low NN weight complexity corresponds to high NN probability in the Bayesian view(e.g.,MacKay,1992;Buntine and Weigend,1991;De Freitas,2003), and to high generalization performance(e.g.,Baum and Haussler,1989),without overfitting the training data.Many methods have been proposed for regularizing NNs,that is,searching for solution-computing, low-complexity SL NNs(Sec.5.6.3)and RL NNs(Sec.6.7).This is closely related to certain UL methods (Sec.4.2,5.6.4).4.4Learning Hierarchical Representations Through Deep SL,UL,RLMany methods of Good Old-Fashioned Artificial Intelligence(GOFAI)(Nilsson,1980)as well as more recent approaches to AI(Russell et al.,1995)and Machine Learning(Mitchell,1997)learn hierarchies of more and more abstract data representations.For example,certain methods of syntactic pattern recog-nition(Fu,1977)such as grammar induction discover hierarchies of formal rules to model observations. The partially(un)supervised Automated Mathematician/EURISKO(Lenat,1983;Lenat and Brown,1984) continually learns concepts by combining previously learnt concepts.Such hierarchical representation learning(Ring,1994;Bengio et al.,2013;Deng and Yu,2014)is also a recurring theme of DL NNs for SL (Sec.5),UL-aided SL(Sec.5.7,5.10,5.15),and hierarchical RL(Sec.6.5).Often,abstract hierarchical representations are natural by-products of data compression(Sec.4.3),e.g.,Sec.5.10.4.5Fast Graphics Processing Units(GPUs)for DL in NNsWhile the previous millennium saw several attempts at creating fast NN-specific hardware(e.g.,Jackel et al.,1990;Faggin,1992;Ramacher et al.,1993;Widrow et al.,1994;Heemskerk,1995;Korkin et al., 1997;Urlbe,1999),and at exploiting standard hardware(e.g.,Anguita et al.,1994;Muller et al.,1995; Anguita and Gomes,1996),the new millennium brought a DL breakthrough in form of cheap,multi-processor graphics cards or GPUs.GPUs are widely used for video games,a huge and competitive market that has driven down hardware prices.GPUs excel at fast matrix and vector multiplications required not only for convincing virtual realities but also for NN training,where they can speed up learning by a factorof50and more.Some of the GPU-based FNN implementations(Sec.5.16-5.19)have greatly contributed to recent successes in contests for pattern recognition(Sec.5.19-5.22),image segmentation(Sec.5.21), and object detection(Sec.5.21-5.22).5Supervised NNs,Some Helped by Unsupervised NNsThe main focus of current practical applications is on Supervised Learning(SL),which has dominated re-cent pattern recognition contests(Sec.5.17-5.22).Several methods,however,use additional Unsupervised Learning(UL)to facilitate SL(Sec.5.7,5.10,5.15).It does make sense to treat SL and UL in the same section:often gradient-based methods,such as BP(Sec.5.5.1),are used to optimize objective functions of both UL and SL,and the boundary between SL and UL may blur,for example,when it comes to time series prediction and sequence classification,e.g.,Sec.5.10,5.12.A historical timeline format will help to arrange subsections on important inspirations and techni-cal contributions(although such a subsection may span a time interval of many years).Sec.5.1briefly mentions early,shallow NN models since the1940s,Sec.5.2additional early neurobiological inspiration relevant for modern Deep Learning(DL).Sec.5.3is about GMDH networks(since1965),perhaps thefirst (feedforward)DL systems.Sec.5.4is about the relatively deep Neocognitron NN(1979)which is similar to certain modern deep FNN architectures,as it combines convolutional NNs(CNNs),weight pattern repli-cation,and winner-take-all(WTA)mechanisms.Sec.5.5uses the notation of Sec.2to compactly describe a central algorithm of DL,namely,backpropagation(BP)for supervised weight-sharing FNNs and RNNs. It also summarizes the history of BP1960-1981and beyond.Sec.5.6describes problems encountered in the late1980s with BP for deep NNs,and mentions several ideas from the previous millennium to overcome them.Sec.5.7discusses afirst hierarchical stack of coupled UL-based Autoencoders(AEs)—this concept resurfaced in the new millennium(Sec.5.15).Sec.5.8is about applying BP to CNNs,which is important for today’s DL applications.Sec.5.9explains BP’s Fundamental DL Problem(of vanishing/exploding gradients)discovered in1991.Sec.5.10explains how a deep RNN stack of1991(the History Compressor) pre-trained by UL helped to solve previously unlearnable DL benchmarks requiring Credit Assignment Paths(CAPs,Sec.3)of depth1000and more.Sec.5.11discusses a particular WTA method called Max-Pooling(MP)important in today’s DL FNNs.Sec.5.12mentions afirst important contest won by SL NNs in1994.Sec.5.13describes a purely supervised DL RNN(Long Short-Term Memory,LSTM)for problems of depth1000and more.Sec.5.14mentions an early contest of2003won by an ensemble of shallow NNs, as well as good pattern recognition results with CNNs and LSTM RNNs(2003).Sec.5.15is mostly about Deep Belief Networks(DBNs,2006)and related stacks of Autoencoders(AEs,Sec.5.7)pre-trained by UL to facilitate BP-based SL.Sec.5.16mentions thefirst BP-trained MPCNNs(2007)and GPU-CNNs(2006). Sec.5.17-5.22focus on official competitions with secret test sets won by(mostly purely supervised)DL NNs since2009,in sequence recognition,image classification,image segmentation,and object detection. Many RNN results depended on LSTM(Sec.5.13);many FNN results depended on GPU-based FNN code developed since2004(Sec.5.16,5.17,5.18,5.19),in particular,GPU-MPCNNs(Sec.5.19).5.11940s and EarlierNN research started in the1940s(e.g.,McCulloch and Pitts,1943;Hebb,1949);compare also later work on learning NNs(Rosenblatt,1958,1962;Widrow and Hoff,1962;Grossberg,1969;Kohonen,1972; von der Malsburg,1973;Narendra and Thathatchar,1974;Willshaw and von der Malsburg,1976;Palm, 1980;Hopfield,1982).In a sense NNs have been around even longer,since early supervised NNs were essentially variants of linear regression methods going back at least to the early1800s(e.g.,Legendre, 1805;Gauss,1809,1821).Early NNs had a maximal CAP depth of1(Sec.3).5.2Around1960:More Neurobiological Inspiration for DLSimple cells and complex cells were found in the cat’s visual cortex(e.g.,Hubel and Wiesel,1962;Wiesel and Hubel,1959).These cellsfire in response to certain properties of visual sensory inputs,such as theorientation of plex cells exhibit more spatial invariance than simple cells.This inspired later deep NN architectures(Sec.5.4)used in certain modern award-winning Deep Learners(Sec.5.19-5.22).5.31965:Deep Networks Based on the Group Method of Data Handling(GMDH) Networks trained by the Group Method of Data Handling(GMDH)(Ivakhnenko and Lapa,1965; Ivakhnenko et al.,1967;Ivakhnenko,1968,1971)were perhaps thefirst DL systems of the Feedforward Multilayer Perceptron type.The units of GMDH nets may have polynomial activation functions imple-menting Kolmogorov-Gabor polynomials(more general than traditional NN activation functions).Given a training set,layers are incrementally grown and trained by regression analysis,then pruned with the help of a separate validation set(using today’s terminology),where Decision Regularisation is used to weed out superfluous units.The numbers of layers and units per layer can be learned in problem-dependent fashion. This is a good example of hierarchical representation learning(Sec.4.4).There have been numerous ap-plications of GMDH-style networks,e.g.(Ikeda et al.,1976;Farlow,1984;Madala and Ivakhnenko,1994; Ivakhnenko,1995;Kondo,1998;Kord´ık et al.,2003;Witczak et al.,2006;Kondo and Ueno,2008).5.41979:Convolution+Weight Replication+Winner-Take-All(WTA)Apart from deep GMDH networks(Sec.5.3),the Neocognitron(Fukushima,1979,1980,2013a)was per-haps thefirst artificial NN that deserved the attribute deep,and thefirst to incorporate the neurophysiolog-ical insights of Sec.5.2.It introduced convolutional NNs(today often called CNNs or convnets),where the(typically rectangular)receptivefield of a convolutional unit with given weight vector is shifted step by step across a2-dimensional array of input values,such as the pixels of an image.The resulting2D array of subsequent activation events of this unit can then provide inputs to higher-level units,and so on.Due to massive weight replication(Sec.2),relatively few parameters may be necessary to describe the behavior of such a convolutional layer.Competition layers have WTA subsets whose maximally active units are the only ones to adopt non-zero activation values.They essentially“down-sample”the competition layer’s input.This helps to create units whose responses are insensitive to small image shifts(compare Sec.5.2).The Neocognitron is very similar to the architecture of modern,contest-winning,purely super-vised,feedforward,gradient-based Deep Learners with alternating convolutional and competition lay-ers(e.g.,Sec.5.19-5.22).Fukushima,however,did not set the weights by supervised backpropagation (Sec.5.5,5.8),but by local un supervised learning rules(e.g.,Fukushima,2013b),or by pre-wiring.In that sense he did not care for the DL problem(Sec.5.9),although his architecture was comparatively deep indeed.He also used Spatial Averaging(Fukushima,1980,2011)instead of Max-Pooling(MP,Sec.5.11), currently a particularly convenient and popular WTA mechanism.Today’s CNN-based DL machines profita lot from later CNN work(e.g.,LeCun et al.,1989;Ranzato et al.,2007)(Sec.5.8,5.16,5.19).5.51960-1981and Beyond:Development of Backpropagation(BP)for NNsThe minimisation of errors through gradient descent(Hadamard,1908)in the parameter space of com-plex,nonlinear,differentiable,multi-stage,NN-related systems has been discussed at least since the early 1960s(e.g.,Kelley,1960;Bryson,1961;Bryson and Denham,1961;Pontryagin et al.,1961;Dreyfus,1962; Wilkinson,1965;Amari,1967;Bryson and Ho,1969;Director and Rohrer,1969;Griewank,2012),ini-tially within the framework of Euler-LaGrange equations in the Calculus of Variations(e.g.,Euler,1744). Steepest descent in such systems can be performed(Bryson,1961;Kelley,1960;Bryson and Ho,1969)by iterating the ancient chain rule(Leibniz,1676;L’Hˆo pital,1696)in Dynamic Programming(DP)style(Bell-man,1957).A simplified derivation of the method uses the chain rule only(Dreyfus,1962).The methods of the1960s were already efficient in the DP sense.However,they backpropagated derivative information through standard Jacobian matrix calculations from one“layer”to the previous one, explicitly addressing neither direct links across several layers nor potential additional efficiency gains due to network sparsity(but perhaps such enhancements seemed obvious to the authors).。

When introducing a notable figure in an English essay,it is essential to provide a comprehensive overview that includes their background,achievements,and impact on society or their field of expertise.Here is a detailed example of how to write an essay introducing a famous person:Title:The Life and Legacy of Albert EinsteinIntroductionAlbert Einstein,born on March14,1879,in Ulm,Germany,is a name synonymous with genius in the world of science.His groundbreaking work in theoretical physics revolutionized our understanding of the universe and laid the foundation for modern physics.This essay aims to explore Einsteins early life,his most significant contributions to science,and his lasting legacy.Early Life and EducationEinsteins childhood was marked by a deep curiosity about the world around him.He showed an early interest in mathematics and physics,often spending hours pondering complex problems.Despite initial struggles in school,Einsteins passion for learning led him to enroll at the Swiss Federal Institute of Technology in Zurich,where he pursued a degree in physics.Contributions to ScienceEinsteins most famous contribution is undoubtedly the theory of relativity,which includes both the special theory of relativity and the general theory of relativity.The special theory,published in1905,challenged the existing notions of space and time, proposing that they are relative to the observers motion.The general theory,published in 1915,expanded on this by introducing the concept of gravity as a curvature of spacetime caused by mass.Another significant contribution was the massenergy equivalence formula,Emc²,which has become a cornerstone of modern physics.This equation demonstrated that mass and energy are interchangeable,a principle that underpins nuclear energy and has profound implications for our understanding of the universe.Impact on SocietyEinsteins theories have had farreaching implications beyond the realm of physics.His work has influenced fields such as cosmology,where the theory of relativity has been used to understand the behavior of black holes and the expansion of the universe. Additionally,his massenergy equivalence formula has been crucial in the development of nuclear technology,both for energy production and weaponry.Einstein was also a vocal advocate for peace and disarmament,particularly in the wake of the atomic bombings of Hiroshima and Nagasaki.His public stance against nuclear weapons and his involvement in the establishment of the Emergency Committee of Atomic Scientists highlight his commitment to using his influence for the betterment of humanity.LegacyAlbert Einstein passed away on April18,1955,but his legacy continues to inspire generations of scientists and thinkers.His theories have been confirmed through numerous experiments and observations,solidifying their place in the annals of scientific discovery.Moreover,Einsteins life serves as a testament to the power of curiosity and the importance of challenging established norms.ConclusionIn conclusion,Albert Einsteins life and work exemplify the spirit of scientific inquiry and the pursuit of knowledge.His theories have not only expanded our understanding of the physical world but have also shaped our cultural and intellectual landscape.As we continue to explore the mysteries of the universe,the legacy of Albert Einstein remains a guiding light for scientists and dreamers alike.This essay provides a structured approach to introducing a famous person,focusing on their life story,contributions,and impact,while also reflecting on their legacy and the broader implications of their work.。

U n i t1T e x t A N e u r o n O v e r l o a d a n d t h e J u g g l i n gP h y s i c i a nUnit 1 Text A神经过载与千头万绪的医生患者经常抱怨自己的医生不会聆听他们的诉说。

虽然可能会有那么几个医生确实充耳不闻，但是大多数医生通情达理，还是能够感同身受的人。

我就纳闷为什么即使这些医生似乎成为批评的牺牲品。

我常常想这个问题的成因是不是就是医生所受的神经过载。

有时我感觉像变戏法，大脑千头万绪，事无巨细，不能挂一漏万。

如果病人冷不丁提个要求，即使所提要求十分中肯，也会让我那内心脆弱的平衡乱作一团，就像井然有序同时演出三台节目的大马戏场突然间崩塌了一样。

有一天，我算过一次常规就诊过程中我脑子里有多少想法在翻腾，试图据此弄清楚为了完满完成一项工作，一个医生的脑海机灵转动，需要处理多少个细节。

奥索里奥夫人56岁，是我的病人。

她有点超重。

她的糖尿病和高血压一直控制良好，恰到好处。

她的胆固醇偏高，但并没有服用任何药物。

她锻炼不够多，最后一次DEXA骨密度检测显示她的骨质变得有点疏松。

尽管她一直没有爽约，按时看病，并能按时做血液化验，但是她形容自己的生活还有压力。

总的说来，她健康良好，在医疗实践中很可能被描述为一个普通患者，并非过于复杂。

以下是整个20分钟看病的过程中我脑海中闪过的念头。

她做了血液化验，这是好事。

血糖好点了。

胆固醇不是很好。

可能需要考虑开始服用他汀类药物。

她的肝酶正常吗？她的体重有点增加。

我需要和她谈谈每天吃五种蔬果、每天步行30分钟的事。

糖尿病：她早上的血糖水平和晚上的比对结果如何？她最近是否和营养师谈过？她是否看过眼科医生？足科医生呢？她的血压还好，但不是很好。

我是不是应该再加一种降血压的药？药片多了是否让她困惑？更好地控制血压的益处和她可能什么药都不吃带来的风险孰重孰轻？骨密度DEXA扫描显示她的骨质有点疏松。

Unit?1?Text?A?神经过载与千头万绪的医生患者经常抱怨自己的医生不会聆听他们的诉说。

虽然可能会有那么几个医生确实充耳不闻，但是大多数医生通情达理，还是能够感同身受的人。

我就纳闷为什么即使这些医生似乎成为批评的牺牲品。

我常常想这个问题的成因是不是就是医生所受的神经过载。

有时我感觉像变戏法，大脑千头万绪，事无巨细，不能挂一漏万。

如果病人冷不丁提个要求，即使所提要求十分中肯，也会让我那内心脆弱的平衡乱作一团，就像井然有序同时演出三台节目的大马戏场突然间崩塌了一样。

?有一天，我算过一次常规就诊过程中我脑子里有多少想法在翻腾，试图据此弄清楚为了完满完成一项工作，一个医生的脑海机灵转动，需要处理多少个细节。

奥索里奥夫人56岁，是我的病人。

她有点超重。

她的糖尿病和高血压一直控制良好，恰到好处。

她的胆固醇偏高，但并没有服用任何药物。

她锻炼不够多，最后一次DEXA骨密度检测显示她的骨质变得有点疏松。

尽管她一直没有爽约，按时看病，并能按时做血液化验，但是她形容自己的生活还有压力。

总的说来，她健康良好，在医疗实践中很可能被描述为一个普通患者，并非过于复杂。

?以下是整个20分钟看病的过程中我脑海中闪过的念头。

?她做了血液化验，这是好事。

?血糖好点了。

胆固醇不是很好。

可能需要考虑开始服用他汀类药物。

她的肝酶正常吗？??她的体重有点增加。

我需要和她谈谈每天吃五种蔬果、每天步行30分钟的事。

??糖尿病：她早上的血糖水平和晚上的比对结果如何？她最近是否和营养师谈过？她是否看过眼科医生？足科医生呢？??她的血压还好，但不是很好。

我是不是应该再加一种降血压的药？药片多了是否让她困惑？更好地控制血压的益处和她可能什么药都不吃带来的风险孰重孰轻？??骨密度DEXA扫描显示她的骨质有点疏松。

我是否应该让她服用二磷酸盐，因为这可以预防骨质疏松症？而我现在又要给她加一种药丸，而这种药需要详细说明。

也许留到下一次再说吧？??她家里的情况怎么样呢？她现在是否有常见的生活压力？亦或她有可能有抑郁症或焦虑症？有没有时间让她做个抑郁问卷调查呢？健康保养：她最后一次乳房X光检查是什么时候做的？子宫颈抹片呢？50岁之后是否做过结肠镜检查？过去10年间她是否注射过破伤风加强疫苗？她是否符合接种肺炎疫苗的条件？奥索里奥夫人打断了我的思路，告诉我过去的几个月里她一直背痛。

神经生物学家英语Neurobiology is a fascinating field that explores the intricate workings of the nervous system, from the molecular level to the complex behaviors it supports. Neurobiologistsare scientists who specialize in understanding the biological basis of the nervous system, which includes the brain, spinal cord, and peripheral nerves.These researchers delve into various aspects of neurobiology, such as the structure and function of neurons, the transmission of signals across synapses, and the development of the nervous system. They also investigate neurological disorders, such as Alzheimer's disease,Parkinson's disease, and multiple sclerosis, seeking to uncover the underlying mechanisms that contribute to these conditions.The study of neurobiology is crucial for advancing our understanding of the brain and its role in cognition, emotion, and behavior. By examining the neural circuits that underlie sensory perception, learning, memory, and decision-making, neurobiologists contribute to the development of treatmentsfor a wide range of neurological and psychiatric disorders.Technological advancements have greatly facilitated the work of neurobiologists. Techniques such as functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and optogenetics allow researchers to observe andmanipulate neural activity in real-time, providing unprecedented insights into the brain's functioning.Moreover, the interdisciplinary nature of neurobiology means that neurobiologists often collaborate with experts in fields such as genetics, psychology, computer science, and engineering. This collaboration fosters innovative approaches to studying the brain and developing new therapies for neurological conditions.In conclusion, neurobiologists play a pivotal role in unraveling the mysteries of the nervous system. Their research not only deepens our understanding of the brain but also has the potential to transform the lives of those affected by neurological disorders. As the field continues to evolve, the contributions of neurobiologists will undoubtedly remain at the forefront of scientific discovery.。