Pietro Vertechi

Pietro Vertechi

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 March 2018 in Nature Communications
Verified
Journal Article
Open access
PDF file
ELEran Lottem
Eran Lottem
DBDhruba Banerjee
Dhruba Banerjee
PVPietro Vertechi
Pietro Vertechi
DSDario Sarra
Dario Sarra
MLMatthijs Oude Lohuis
Matthijs Oude Lohuis
ZMZachary F. Mainen
Zachary F. Mainen
The neuromodulator serotonin (5-HT) has been implicated in a variety of functions that involve patience or impulse control. Many of these effects are consistent with a long-standing theory that 5-HT promotes behavioral inhibition, a motivational bias favoring passive over active behaviors. To further test this idea, we studied the impact of 5-HT in a probabilistic foraging task, in which mice must learn the statistics of the environment and infer when to leave a depleted foraging site for the next. Critically, mice were required to actively nose-poke in order to exploit a given site. We show that optogenetic activation of 5-HT neurons in the dorsal raphe nucleus increases the willingness of mice to actively attempt to exploit a reward site before giving up. These results indicate that behavioral inhibition is not an adequate description of 5-HT function and suggest that a unified account must be based on a higher-order function.

16 March 2020 in PLOS Computational Biology
Verified
Journal Article
Open access
PDF file
WBWieland Brendel
Wieland Brendel
RBRalph Bourdoukan
Ralph Bourdoukan
PVPietro Vertechi
Pietro Vertechi
CMChristian K. Machens
Christian K. Machens
SDSophie Denève
Sophie Denève
Networks based on coordinated spike coding can encode information with high efficiency in the spike trains of individual neurons. These networks exhibit single-neuron variability and tuning curves as typically observed in cortex, but paradoxically coincide with a precise, non-redundant spike-based population code. However, it has remained unclear whether the specific synaptic connectivities required in these networks can be learnt with local learning rules. Here, we show how to learn the required architecture. Use coding efficiency as an objective, we derive spike-timing-dependent learning rules for a recurrent neural network, and we provide exact solutions for the networks’ convergence to an optimal state. As a result, we deduce an entire network from its input distribution and a firing cost. After learning, basic biophysical quantities such as voltages, firing thresholds, excitation, inhibition, or spikes acquire precise functional interpretations. Spiking neural networks can encode information with high efficiency in the spike trains of individual neurons if the synaptic weights between neurons are set to specific, optimal values. In this regime, the networks exhibit irregular spike trains, high trial-to-trial variability, and stimulus tuning as typically observed in cortex. The strong variability on the level of single neurons paradoxically coincides with a precise, non-redundant spike-based population code. However, it has remained unclear whether the specific synaptic connectivities required in these spiking networks can be learnt with local learning rules. In this study, we show how the required architecture can be learnt. We derive local and biophysically plausible learning rules for a recurrent neural networks from first principles. We show both mathematically and using numerical simulations that these learning rules drive the networks into the optimal state, and we show that the optimal state is governed by the statistics of the input signals. After learning, the voltages of individual neurons can be interpreted as measuring the instantaneous error of the code, given by the error between the desired output signal and the actual output signal.

29 November 2021 in Mathematics
Verified
Journal Article
Open access
PDF file
MBMattia Bergomi
Mattia Bergomi
Massimo FerriMassimo Ferri
Massimo Ferri
PVPietro Vertechi
Pietro Vertechi
LZLorenzo Zuffi
Lorenzo Zuffi
Persistent homology enables fast and computable comparison of topological objects. We give some instances of a recent extension of the theory of persistence, guaranteeing robustness and computability for relevant data types, like simple graphs and digraphs. We focus on categorical persistence functions that allow us to study in full generality strong kinds of connectedness—clique communities, k-vertex, and k-edge connectedness—directly on simple graphs and strong connectedness in digraphs.

26 June 2023 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
PVPietro Vertechi
Pietro Vertechi
MBMattia Bergomi
Mattia Bergomi
We provide a unifying framework where artificial neural networks and their architectures can be formally described as particular cases of a general mathematical construction---machines of finite depth. Unlike neural networks, machines have a precise definition, from which several properties follow naturally. Machines of finite depth are modular (they can be combined), efficiently computable, and differentiable. The backward pass of a machine is again a machine and can be computed without overhead using the same procedure as the forward pass. We prove this statement theoretically and practically via a unified implementation that generalizes several classical architectures---dense, convolutional, and recurrent neural networks with a rich shortcut structure---and their respective backpropagation rules.

Verified
Journal Article
Open access
PDF file
MBMattia Bergomi
Mattia Bergomi
Massimo FerriMassimo Ferri
Massimo Ferri
AMAlessandro Mella
Alessandro Mella
PVPietro Vertechi
Pietro Vertechi
Artificial neural networks can learn complex, salient data features to achieve a given task. On the opposite end of the spectrum, mathematically grounded methods such as topological data analysis allow users to design analysis pipelines fully aware of data constraints and symmetries. We introduce an original class of neural network layers based on a generalization of topological persistence. The proposed persistence-based layers allow the users to encode specific data properties (e.g., equivariance) easily. Additionally, these layers can be trained through standard optimization procedures (backpropagation) and composed with classical layers. We test the performance of generalized persistence-based layers as pooling operators in convolutional neural networks for image classification on the MNIST, Fashion-MNIST and CIFAR-10 datasets.

28 December 2022 in arXiv
Verified
Preprint
Open access
PDF file
MBMattia Bergomi
Mattia Bergomi
Massimo FerriMassimo Ferri
Massimo Ferri
AMAlessandro Mella
Alessandro Mella
PVPietro Vertechi
Pietro Vertechi
Artificial neural networks can learn complex, salient data features to achieve a given task. On the opposite end of the spectrum, mathematically grounded methods such as topological data analysis allow users to design analysis pipelines fully aware of data constraints and symmetries. We introduce a class of persistence-based neural network layers. Persistence-based layers allow the users to easily inject knowledge about symmetries (equivariance) respected by the data, are equipped with learnable weights, and can be composed with state-of-the-art neural architectures.