Humans use log-compressed number lines to represent different quantities, including elapsed time, traveled distance, numerosity, sound frequency, etc. Inspired by recent cognitive science and computational neuroscience work, we developed a neural network that learns to construct log-compressed number lines. The network computes a discrete approximation of a real-domain Laplace transform using an RNN with analytically derived weights giving rise to a log-compressed timeline of the past. The network learns to extract latent variables from the input and uses them for global modulation of the recurrent weights turning a timeline into a number line over relevant dimensions. The number line representation greatly simplifies learning on a set of problems that require learning associations in different spaces - problems that humans can typically solve easily. This approach illustrates how combining deep learning with cognitive models can result in systems that learn to represent latent variables in a brain-like manner and exhibit human-like behavior manifested through Weber-Fechner law.

Human learners can readily understand speech, or a melody, when it is presented slower or faster than usual. This paper presents a deep CNN (SITHCon) that uses a logarithmically compressed temporal representation at each level. Because rescaling the time of the input results in a translation of $\log$ time, and because the output of the convolution is invariant to translations, this network can generalize to out-of-sample data that are temporal rescalings of a learned pattern. We compare the performance of SITHCon to a Temporal Convolution Network (TCN) on classification and regression problems with both univariate and multivariate time series. We find that SITHCon, unlike TCN, generalizes robustly over rescalings of about an order of magnitude. Moreover, we show that the network can generalize over exponentially large scales without retraining the weights simply by extending the range of the logarithmically-compressed temporal memory.

In recent years, it has become clear that the brain maintains a temporal memory of recent events stretching far into the past. This letter presents a neurally inspired algorithm to use a scale-invariant temporal representation of the past to predict a scale-invariant future. The result is a scale-invariant estimate of future events as a function of the time at which they are expected to occur. The algorithm is time-local, with credit assigned to the present event by observing how it affects the prediction of the future. To illustrate the potential utility of this approach, we test the model on simultaneous renewal processes with different timescales. The algorithm scales well on these problems despite the fact that the number of states needed to describe them as a Markov process grows exponentially.

Extracting temporal relationships over a range of scales is a hallmark ofhuman perception and cognition---and thus it is a critical feature of machinelearning applied to real-world problems. Neural networks are either plaguedby the exploding/vanishing gradient problem in recurrent neural networks(RNNs) or must adjust their parameters to learn the relevant time scales(e.g., in LSTMs). This paper introduces DeepSITH, a deep network comprisingbiologically-inspired Scale-Invariant Temporal History (SITH) modules inseries with dense connections between layers. Each SITH module is simply aset of time cells coding what happened when with a geometrically-spaced set oftime lags. The dense connections between layers change the definition of whatfrom one layer to the next. The geometric series of time lags implies thatthe network codes time on a logarithmic scale, enabling DeepSITH network tolearn problems requiring memory over a wide range of time scales. We compareDeepSITH to LSTMs and other recent RNNs on several time series prediction anddecoding tasks. DeepSITH achieves results comparable to state-of-the-artperformance on these problems and continues to perform well even as the delaysare increased.

Sequential neural activity has been observed in many parts of the brain and has been proposed as a neural mechanism for memory. The natural world expresses temporal relationships at a wide range of scales. Because we cannot know the relevant scales a priori, it is desirable that memory, and thus the generated sequences, is scale invariant. Although recurrent neural network models have been proposed as a mechanism for generating sequences, the requirements for scale-invariant sequences are not known. This letter reports the constraints that enable a linear recurrent neural network model to generate scale-invariant sequential activity. A straightforward eigendecomposition analysis results in two independent conditions that are required for scale invariance for connectivity matrices with real, distinct eigenvalues. First, the eigenvalues of the network must be geometrically spaced. Second, the eigenvectors must be related to one another via translation. These constraints are easily generalizable for matrices that have complex and distinct eigenvalues. Analogous albeit less compact constraints hold for matrices with degenerate eigenvalues. These constraints, along with considerations on initial conditions, provide a general recipe to build linear recurrent neural networks that support scale-invariant sequential activity.

Natural learners must compute an estimate of future outcomes that follow from a stimulus in continuous time. Widely used reinforcement learning algorithms discretize continuous time and estimate either transition functions from one step to the next (model-based algorithms) or a scalar value of exponentially discounted future reward using the Bellman equation (model-free algorithms). An important drawback of model-based algorithms is that computational cost grows linearly with the amount of time to be simulated. An important drawback of model-free algorithms is the need to select a timescale required for exponential discounting. We present a computational mechanism, developed based on work in psychology and neuroscience, for computing a scale-invariant timeline of future outcomes. This mechanism efficiently computes an estimate of inputs as a function of future time on a logarithmically compressed scale and can be used to generate a scale-invariant power-law-discounted estimate of expected future reward. The representation of future time retains information about what will happen when. The entire timeline can be constructed in a single parallel operation that generates concrete behavioral and neural predictions. This computational mechanism could be incorporated into future reinforcement learning algorithms.

Predicting the timing and order of future events is an essential feature of cognition in higher life forms. We propose a neural mechanism to nondestructively translate the current state of spatiotemporal memory into the future, so as to construct an ordered set of future predictions almost instantaneously. We hypothesize that within each cycle of hippocampal theta oscillations, the memory state is swept through a range of translations to yield an ordered set of future predictions through modulations in synaptic connections. Theoretically, we operationalize critical neurobiological findings from hippocampal physiology in terms of neural network equations representing spatiotemporal memory. Combined with constraints based on physical principles requiring scale invariance and coherence in translation across memory nodes, the proposition results in Weber-Fechner spacing for the representation of both past (memory) and future (prediction) timelines. We show that the phenomenon of phase precession of neurons in the hippocampus and ventral striatum correspond to the cognitive act of future prediction.

We propose a principled way to construct an internal representation of the temporal stimulus history leading up to the present moment. A set of leaky integrators performs a Laplace transform on the stimulus function, and a linear operator approximates the inversion of the Laplace transform. The result is a representation of stimulus history that retains information about the temporal sequence of stimuli. This procedure naturally represents more recent stimuli more accurately than less recent stimuli; the decrement in accuracy is precisely scale invariant. This procedure also yields time cells that fire at specific latencies following the stimulus with a scale-invariant temporal spread. Combined with a simple associative memory, this representation gives rise to a moment-to-moment prediction that is also scale invariant in time. We propose that this scale-invariant representation of temporal stimulus history could serve as an underlying representation accessible to higher-level behavioral and cognitive mechanisms. In order to illustrate the potential utility of this scale-invariant representation in a variety of fields, we sketch applications using minimal performance functions to problems in classical conditioning, interval timing, scale-invariant learning in autoshaping, and the persistence of the recency effect in episodic memory across timescales.

Semantic memory refers to our knowledge of facts and relationships between con- cepts. A successful semantic memory depends on inferring relationships between items that are not explicitly taught. Recent mathematical modeling of episodic memory argues that episodic recall relies on retrieval of a gradually-changing rep- resentation of temporal context. We show that retrieved context enables the de- velopment of a global memory space that reflects relationships between all items that have been previously learned. When newly-learned information is integrated into this structure, it is placed in some relationship to all other items, even if that relationship has not been explicitly learned. We demonstrate this effect for global semantic structures shaped topologically as a ring, and as a two-dimensional sheet. We also examined the utility of this learning algorithm for learning a more realistic semantic space by training it on a large pool of synonym pairs. Retrieved context enabled the model to “infer” relationships between synonym pairs that had not yet been presented.