The dynamic membrane potential threshold, as one of the essential properties of a biological neuron, is a spontaneous regulation mechanism that maintains neuronal homeostasis, i.e., the constant overall spiking firing rate of a neuron. As such, the neuron firing rate is regulated by a dynamic spiking threshold, which has been extensively studied in biology. Existing work in the machine learning community does not employ bioinspired spiking threshold schemes. This work aims at bridging this gap by introducing a novel bioinspired dynamic energy-temporal threshold (BDETT) scheme for spiking neural networks (SNNs). The proposed BDETT scheme mirrors two bioplausible observations: a dynamic threshold has 1) a positive correlation with the average membrane potential and 2) a negative correlation with the preceding rate of depolarization. We validate the effectiveness of the proposed BDETT on robot obstacle avoidance and continuous control tasks under both normal conditions and various degraded conditions, including noisy observations, weights, and dynamic environments. We find that the BDETT outperforms existing static and heuristic threshold approaches by significant margins in all tested conditions, and we confirm that the proposed bioinspired dynamic threshold scheme offers homeostasis to SNNs in complex real-world tasks.

A convolutional neural network strongly robust to adversarial perturbations at reasonable computational and performance cost has not yet been demonstrated. The primate visual ventral stream seems to be robust to small perturbations in visual stimuli but the underlying mechanisms that give rise to this robust perception are not understood. In this work, we investigate the role of two biologically plausible mechanisms in adversarial robustness. We demonstrate that the non-uniform sampling performed by the primate retina and the presence of multiple receptive fields with a range of receptive field sizes at each eccentricity improve the robustness of neural networks to small adversarial perturbations. We verify that these two mechanisms do not suffer from gradient obfuscation and study their contribution to adversarial robustness through ablation studies.

Nature has a way of making complex shapes from a set of simple growth rules. The curve of a petal, the swoop of a branch, even the contours of our face are shaped by these processes. What if we could unlock those rules and reverse engineer nature's ability to grow an infinitely diverse array of shapes? Scientists from Harvard's Wyss Institute for Biologically Inspired Engineering and the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) have done just that. In a paper published in the Proceedings of the National Academy of Sciences, the team demonstrates a technique to grow any target shape from any starting shape.

In the Google paper, the authors enumerate many risk factors, design patterns, and anti-patterns to needs to be taken into consideration in an architecture. These include design patterns such as: boundary erosion, entanglement, hidden feedback loops, undeclared consumers, data dependencies and changes in the external world. By contrast, Deep Learning systems (applies equally to machine learning), code is created from training data. A recent paper from the folks at Berkeley are exploring the requirements for building these new kinds of systems (see: "Real-Time Machine Learning: The Missing Pieces").

BICA Labs Laboratories for Biologically Inspired Cognitive Architectures 2. Blockchain: Distributed Ledger Technology (DLT) TRUSTED PERSISTENT DATA RECORDS INSIDE TRUSTLESS ENVIRONMENTS WITHOUT CENTRAL GOVERNANCE SECURED BY ECONOMIC INCENTIVES 4. Distributed database (ledger) Distributed computations (state changes), Turing-complete OR incomplete Peer-to-peer mesh network Cryptographically secured TECHNOLOGY ECONOMICS BLOCKCHAIN Multiagent economy Game theory Free open market Non-state decentralized economies linked to particular types of resources or businesses 6. NO CENTRALIZATION 7. NO DATA OLIGOPOLY 9. Blockchain tech intro 10. Important Blockchains Bitcoin Ripple Ethereum ZCashDashNEM Most popular cryptocurrency & source code base Value transfer network Smart contracts Proof of Importance Governance model Zero-knowledge 11. Most important ready-to-go DLTs cores Bitcoin Core Graphene Scorex Tendermint Ripple / Stellar Language C/C C Scala Go C Consensus PoW dPoS PoW, 2x PoS PoS Blockchains Bitcoin, Dash, Litecoin, ... BitShares, Steem, Golos – (Waves experiments) Cosmos (under dev) Ripple, Stellar, Infra (e-Auction) " " Most proved Blazingly fast Modular PoS Fast & proved "–" Hard to understand Complex model Limited functionality, no real-world impl Immature Complex model 16. Languages for working with DLTs cores C/C -- Bitcoin, Graphene, Ripple: performance Go -- Ethereum, Tehndermint Python -- Ethereum experiments with blockchain Rust -- Ethereum, Bitcoin: performance efficient code Scala -- Scorex (fast blockchain prototyping) Java -- NEM JavaScript (Angular, React): UI & APIs 17. Meta-languages for smart contracts Solidity: JavaScript-like Serpent: Python-like Viper: Python-like (Serpent 2.0) 18. Blockchain & AI Technical Synergy 19. Civilization 4.0 key factors Quantum Computing Generic Artificial Intelligence Transhumanism Life extension Cyborgization Cosmic Expansion SINGULARITY 31.

A recent paper suggests that Deep Neural Networks can be protected from gradient-based adversarial perturbations by driving the network activations into a highly saturated regime. Here we analyse such saturated networks and show that the attacks fail due to numerical limitations in the gradient computations. A simple stabilisation of the gradient estimates enables successful and efficient attacks. Thus, it has yet to be shown that the robustness observed in highly saturated networks is not simply due to numerical limitations.

Inspired by biophysical principles underlying nonlinear dendritic computation in neural circuits, we develop a scheme to train deep neural networks to make them robust to adversarial attacks. Our scheme generates highly nonlinear, saturated neural networks that achieve state of the art performance on gradient based adversarial examples on MNIST, despite never being exposed to adversarially chosen examples during training. Moreover, these networks exhibit unprecedented robustness to targeted, iterative schemes for generating adversarial examples, including second-order methods. We further identify principles governing how these networks achieve their robustness, drawing on methods from information geometry. We find these networks progressively create highly flat and compressed internal representations that are sensitive to very few input dimensions, while still solving the task. Moreover, they employ highly kurtotic weight distributions, also found in the brain, and we demonstrate how such kurtosis can protect even linear classifiers from adversarial attack.

Much AI research on computational sustainability has focused on monitoring, modeling, analysis, and optimization of existing systems and processes. In this article, we present another exciting and promising paradigm for AI research on computational sustainability that emphasizes design of new systems and processes, and, in particular, on biologically inspired design. We first characterize biologically inspired design, then examine its relationship with environmental sustainability, next present a computational model of the process of biologically inspired design, and finally describe a few computational systems for supporting biologically inspired design practice.

Sustainable design is as an important movement in design. Biologically inspired design is a major paradigm for sustainable design. In this paper, we analyze a corpus of biologically inspired design projects in terms of sustainability. We then describe a case study of analogical design of a fog harvesting net, and abstract from it the patterns of Hydrophobia and Hydrophilia. We indicate how these two function-mechanism design patterns occur in several design projects in our corpus. This analysis indicates how biologically inspired sustainable design can be analyzed in terms of cross-domain analogical transfer of design patterns.

This extended abstract outlines my invited keynote presentation of the recent work on neuromorphic networks ("CrossNets") based on hybrid CMOS/nanoelectronic ("CMOL") circuits, in the space-saving Q/A format.