Perceptrons

By Dan A. Simovici. 2021

817 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Model Architectures

Citations in the corpus

Inferring Algorithmic Patterns with Stack-Augmented Recurrent Nets (2015, background): “Another example is the parity problem introduced by Minsky & Seymour (1969): it demonstrates that while a single non-linearity is sufﬁcient to represent any function, it is not guaranteed to represent it efﬁciently, and in some cases can even require exponentially many more parameters (and thus,…”
Deep learning in neural networks: An overview (2014, neutral): “…on learning in multilayer NN-like systems (see also Section 5.3 on deep nonlinear nets since 1965), it seems surprising in hindsight that a book (Minsky & Papert, 1969) on the limitations of simple linear perceptrons with a single layer (Section 5.1) discouraged some researchers from further…”
Approximate Nearest Neighbor: Towards Removing the Curse of Dimensionality (2012, background): “Originally posed in the 1960s by Minsky and Papert ([MP69], pp.”
On the mathematical foundations of learning (2001, methodology): “We mention as examples the theory of Neural Networks going back to McCulloch and Pitts [25] and Minsky and Papert [27], the PAC learning of Valiant [40], Statistical Learning Theory as developed by Vapnik [42], and the use of reproducing kernels as in [17] among many other mathematical developments.”
From machine learning to machine reasoning (2011, background): “Minsky and Papert (1969) have shown that simple cognitive tasks cannot be implemented using linear threshold functions but require multiple layers of computation.”
Deep Roots: Improving CNN Efficiency with Hierarchical Filter Groups (2016, neutral): “Interestingly, this is in stark contrast to what we understand of biological neural networks, where we see “highly evolved arrangements of smaller, specialized networks which are interconnected in very speciﬁc ways” [8].”
Detecting near-duplicates for web crawling (2007, methodology): “A generalized version of the Hamming Distance Problem was first proposed by Minsky and Papert [44]: Given a set of n f -bit strings (chosen by an adversary), and a string F , the goal is to identify strings in the set which differ from F in at most d bit-positions.”
Learning DNF in time 2 Õ(n 1/3 ) . (2001, background): “In their well known 1968 book Perceptrons, Minsky and Papert studied some computational aspects of polynomial threshold functions from an Artificial Intelligence perspective [37].”
Learning Real-World Stimuli in a Neural Network with Spike-Driven Synaptic Dynamics (2007, background): “This is an indication that the currently presented pattern has already been learned and that it is not necessary to change the synapses further (stop-learning condition, as in the case of the perceptron learning rule: Rosenblatt, 1958; Block, 1962; Minsky & Papert, 1969).”
Quo vadis artificial intelligence? (2022, background): “In 1969, Minsky and Papert published the Perceptrons [29], which pointed out the significant flaws of perceptron that perceptron can only solving linear problems.”
Geometry of the Restricted Boltzmann Machine (2009, background): “The RBM model grew out of work on artificial neurons modeled as linear threshold functions [24, 28], and we pause our geometric discussion to make a few remarks about these functions and the types of inference functions that our model can represent.”
The unreasonable effectiveness of deep learning in artificial intelligence (2020, methodology): “3) were dashed by Marvin Minsky and Seymour Papert, who showed in their book Perceptrons (8) that a perceptron can only represent categories that are linearly separable in weight space.”
Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network (2018, background): “Subsequently, it has been successfully applied to recurrent networks by taking advantage of the fact that for every recurrent network there exists an equivalent feedforward network with identical behavior for a finite number of steps [42], [43], [30].”
A causal-model theory of conceptual representation and categorization. (2003, neutral): “…within a similarity-based framework, one might postulate the existence of second-order features (Gluck & Bower, 1988; Hayes-Roth & Hayes-Roth, 1977; Minsky & Papert, 1988; Neumann, 1974; Reitman & Bower, 1973; Rumelhart , Hinton, & Williams, 1986) that (a) encode the confirmation or violation of…”
Randomness in neural networks: an overview (2017, methodology): “”(32) This aspect was at the base of the fundamental historical controversy surrounding this model.(33) Later modifications to the perceptron came closer to the RW-FNN as described previously, most notably the variation of the ‘pocket algorithm’ proposed by Gallant and Smith in 1987.”
The polynomial method in circuit complexity (1993, methodology): “***** Minsky and Papert [39] also proved that the AND of two MAJORITYs is not represented by the sign of a degree-O(1) polynomial.”
Fast search in Hamming space with multi-index hashing (2012, methodology): “The 1- NN problem in Hamming space was called the Best Match problem by Minsky and Papert [14].”
Complexity measures and decision tree complexity: a survey (2002, methodology): “Below we sketch the method of symmetrization, due to Minsky and Papert [28] (see also [4, Section 4]).”
Neural networks for pattern recognition (1995, background): “68) is guaranteed to find a solution in a finite number of steps (Rosenblatt, 1962; Block, 1962; Nilsson, 1965; Minsky and Papert, 1969; Duda and Hart, 1973; Hand, 1981; Arbib, 1987; Hertz et al., 1991).”