ライフサイエンスコーパス: backpropagation

1 r signals required by strict formulations of backpropagation.

2 transpose operations required in traditional backpropagation.

3 mplex combines epistemic autonomy with error backpropagation.

4 ade dendrites and dendritic spines by active backpropagation.

5 t onset of somatic AP was produced solely by backpropagation.

6 c potentials and facilitate action potential backpropagation.

7 nts are consistent with active, not passive, backpropagation.

8 t require axonal action potential firing and backpropagation.

9 travel surveys are used as ground truth for backpropagation.

10 S to increase neuronal gain and the speed of backpropagation.

11 ells being diagnostic to be calculated using backpropagation.

12 owing for the calculation of gradients using backpropagation.

13 t parameters are optimized through a revised backpropagation.

14 that this distinct mechanism, in contrast to backpropagation, (1) underlies learning in a well-establ

15 to solve classification tasks using "in situ backpropagation," a photonic analog of the most popular

16 In machine learning, the backpropagation algorithm assigns blame by multiplying e

17 This study presents a neuromorphic, spiking backpropagation algorithm based on synfire-gated dynamic

18 ive power of this modeling framework and the backpropagation algorithm for setting the parameters.

19 iking neural networks (SNNs) using the error backpropagation algorithm has made significant progress

20 y believed that end-to-end training with the backpropagation algorithm is essential for learning good

21 The backpropagation algorithm learns quickly by computing sy

22 edforward networks trained end-to-end with a backpropagation algorithm on simple tasks.

23 The backpropagation algorithm solves this problem in deep ar

24 g Neural Network implementation of the exact backpropagation algorithm that is fully on-chip without

25 classical conditioning tasks using the error backpropagation algorithm to guide learning.

26 te structure in large data sets by using the backpropagation algorithm to indicate how a machine shou

27 so far(10-22) have been unable to apply the backpropagation algorithm to train unconventional novel

28 arge models with the same performance as the backpropagation algorithm widely used in deep learning t

29 , the workhorse of modern deep learning, the backpropagation algorithm, has proven difficult to trans

30 al neural network-based model trained by the backpropagation algorithm.

31 e implemented in a network without using the backpropagation algorithm.

32 In addition, this backpropagation also results in an unusually high rate o

33 ral networks faced a similar challenge until backpropagation and automatic differentiation transforme

34 fire tonically resulting in action potential backpropagation and dendritic Ca(2+) influx.

35 the influence of apical length on dendritic backpropagation and excitability, based on a Na(+) chann

36 ing algorithms, such as k-nearest neighbors, backpropagation and probabilistic neural networks, often

37 of Kv4.2 subunits, regulate action potential backpropagation and the induction of specific forms of s

38 We report enhanced backpropagation and theta resonance and decreased summat

39 ring rates reach 40 Hz (activity-independent backpropagation); and (3) do not exhibit signs of a 'cal

40 activation and loss functions and with both backpropagation- and Hebbian-based learning rules.

41 Backpropagation Applied to Handwritten Zip Code Recognit

42 ing decision trees, support vector machines, backpropagation artificial neural networks, extreme grad

43 hat FSI was associated with action potential backpropagation (bAP) and a supralinear increase in dend

44 To train PNNs, backpropagation-based and backpropagation-free approache

45 fastISM reduces the gap between backpropagation-based feature attribution methods and IS

46 utation effects from Integrated Gradients, a backpropagation-based feature attribution, and character

47 logical cognition can be achieved with error backpropagation-based learning.

48 It far surpasses the runtime of backpropagation-based methods on multi-output architectu

49 tudy on the meteorological data and types of backpropagation (BP) algorithms used to train and develo

50 that average obtained from the conventional backpropagation (BP) method can hardly overcome 0.35 and

51 After backpropagation (BP) training, the ANN model was able to

52 Despite backpropagation (BP)-based training algorithms being the

53 (10 mM) had no effect on activity-dependent backpropagation but blocked the effect of CCh.

54 Instead, recurrent networks trained with backpropagation capture the time-encoding properties and

55 In association with the enhancement of spike backpropagation, CCh increased the amplitude and duratio

56 rd Backpropagation (FFBP) and Cascadeforward Backpropagation (CFBP), were developed and trained using

57 o dopamine, the pulses of GABA prohibited AP backpropagation distally from the application site, even

58 es can likely be explained by differences in backpropagation efficiency, arising from the specific co

59 Two ANN architectures, Feedforward Backpropagation (FFBP) and Cascadeforward Backpropagatio

60 ward-propagating light and simulated in situ backpropagation for 64-port photonic neural networks tra

61 ransforming data representations, as well as backpropagation for model finetuning, saliency map compu

62 th a smooth constitutive update that enables backpropagation for the NN training.

63 To train PNNs, backpropagation-based and backpropagation-free approaches are now being explored.

64 oreover, we demonstrated that AGOP, which is backpropagation-free, enabled feature learning in machin

65 Yet training ANN using error backpropagation has created the current revolution in ar

66 The advantages of backpropagation have made it the de facto training metho

67 Non-decremental, non-dichotomous backpropagation in AOB primary dendrites ensures fast, r

68 ar dendritic computation would support error backpropagation in biological neurons.

69 ty-dependent attenuation of action-potential backpropagation in current-clamp simulations of a CA1 py

70 itic recording data means that the extent of backpropagation in thalamocortical (TC) and thalamic ret

71 alcium dynamics during action potential (AP) backpropagation in three types of V1 supragranular inter

72 Na(V)1.2 loss reduced action potential backpropagation into dendrites, impairing synaptic plast

73 uence synaptic potentiation following active backpropagation into dendrites.

74 tion and undergo strong attenuation in their backpropagation into the dendrites (length constant, 76

75 or local synaptic stimulation, and the rapid backpropagation into the dendritic arbor depended upon v

76 potential repolarization, repetitive firing, backpropagation (into dendrites) of action potentials, a

77 At the same time, the traditional form of backpropagation is biologically implausible.

78 Backpropagation is limited, however, and action potentia

79 Backpropagation is widely used to train artificial neura

80 We implement the Levenberg Marquardt backpropagation (LMB) algorithm for investigating an inn

81 The backpropagation method has enabled transformative uses o

82 mong the various training schemes, the error backpropagation method that directly uses the firing tim

83 nberg-Marquardt, quasi-Newton, and resilient backpropagation methods are employed to train the ANN.

84 phic findings and patient age, a three-layer backpropagation network was developed to predict whether

85 the four classifiers are then fused using a backpropagation neural network (BNN) to diagnose each re

86 analyzed with multivariate statistics and a backpropagation neural network (NN).

87 rs evaluated the performance of feed-forward backpropagation neural networks in predicting rapid prog

88 f the spectra using the multivariate methods backpropagation neural networks, decision tree, adaboost

89 ting the frequency of repetitive firing, the backpropagation of action potential into dendrites, and

90 onsiders nonlinear dendritic integration and backpropagation of action potentials from the soma to th

91 tion of synaptic input or the initiation and backpropagation of action potentials in a branch-selecti

92 ing expression of Nav1.2 channels attenuates backpropagation of action potentials into dendrites of c

93 s to an increase in synaptic integration and backpropagation of action potentials into distal dendrit

94 The backpropagation of action potentials into the dendritic

95 ppear at odds with the unusually weak active backpropagation of action potentials into the soma and d

96 h plasticity between LTD and LTP by boosting backpropagation of action potentials.

97 but is thought to rely in part on dendritic backpropagation of action potentials.

98 n, Na(V)1.2 at the proximal AIS promotes the backpropagation of APs to the soma.

99 Backpropagation of autonomously generated APs was reliab

100 the dendritic arbor, calcium signals during backpropagation of both single APs and AP trains were re

101 of p-hydroxyphenacyl (pHP) GABA demonstrates backpropagation of GABAAR-mediated depolarizations from

102 Backpropagation of somatic action potentials generated S

103 ials in the axon initial segment followed by backpropagation of these spikes throughout the neuron re

104 this computation,(1)(,)(2) facilitating the backpropagation of value from the predicted reward to th

105 works have reinvigorated interest in whether backpropagation offers insights for understanding learni

106 rs of neurons and performs as effectively as backpropagation on a variety of tasks.

107 opamine is known to inhibit action potential backpropagation, our experiments revealed an unexpected

108 Some biological models of backpropagation rely on feedback projections that are sy

109 site of AP generation and the true extent of backpropagation remain unknown.

110 and action potential output caused by spike backpropagation results in the appearance of high spike

111 terpreting deep survival models using a risk backpropagation technique.

112 tal deep-learning models primarily relies on backpropagation that is unsuitable for physical implemen

113 xclusively controls I(NaP) generation and AP backpropagation, thereby playing a prominent role in syn

114 Our approach combines backpropagation through time with cascade learning, enab

115 introduce the mechanical analogue of in situ backpropagation to enable highly efficient training of m

116 ElastNet uses backpropagation to learn the hidden elasticity of object

117 nity, and we experimentally demonstrate this backpropagation to obtain gradient with high precision.

118 called physics-aware training, that applies backpropagation to train controllable physical systems.

119 ng between inner layers is scrambled because backpropagation training does not require perceptrons to

120 Simulations of backpropagation using the device properties reach ideal

121 This spatial control of AP backpropagation was mediated by Ia-type potassium curren

122 Backpropagation was reliant on Na+ channels: in 1 microm

123 med that credit assignment is best solved by backpropagation, which is also the foundation of modern

124 s-aware training combines the scalability of backpropagation with the automatic mitigation of imperfe

125 and (iii) can be specified and trained using backpropagation with the same ease-of-use as contemporar