Constructing custom thermodynamics using deep learning.

One of the most exciting applications of artificial intelligence is automated scientific discovery based on previously amassed data, coupled with restrictions provided by known physical principles, including symmetries and conservation laws. Such automated hypothesis creation and verification can assist scientists in studying complex phenomena, where traditional physical intuition may fail. Here we develop a platform based on a generalized Onsager principle to learn macroscopic dynamical descriptions of arbitrary stochastic dissipative systems directly from observations of their microscopic trajectories. Our method simultaneously constructs reduced thermodynamic coordinates and interprets the dynamics on these coordinates. We demonstrate its effectiveness by studying theoretically and validating experimentally the stretching of long polymer chains in an externally applied field. Specifically, we learn three interpretable thermodynamic coordinates and build a dynamical landscape of polymer stretching, including the identification of stable and transition states and the control of the stretching rate. Our general methodology can be used to address a wide range of scientific and technological applications.

Automated electrokinetic stretcher for manipulating nanomaterials.

In this work, we present an automated platform for trapping and stretching individual micro- and nanoscale objects in solution using electrokinetic forces. The platform can trap objects at the stagnation point of a planar elongational electrokinetic field for long time scales, as demonstrated by the trapping of <100 nm polystyrene beads and DNA molecules for minutes, with a standard deviation in displacement from the trap center <1 µm. This capability enables the stretching of deformable nanoscale objects in a high-throughput fashion, as illustrated by the stretching of more than 400 DNA molecules within ∼4 hours. The flexibility of the electrokinetic stretcher opens up numerous possibilities for complex manipulation, with sequential stretching of a molecule at different voltages and multiple stretch-relaxation cycles of the same molecule as examples. The platform described provides an automated, high-throughput method to track and manipulate objects for real-time studies of micro- and nanoscale systems.

Towards Intelligently Designed Evolvable Processors.

Evolution-in-Materio is a computational paradigm in which an algorithm reconfigures a material's properties to achieve a specific computational function. This article addresses the question of how successful and well performing Evolution-in-Materio processors can be designed through the selection of nanomaterials and an evolutionary algorithm for a target application. A physical model of a nanomaterial network is developed which allows for both randomness, and the possibility of Ohmic and non-Ohmic conduction, that are characteristic of such materials. These differing networks are then exploited by differential evolution, which optimises several configuration parameters (e.g., configuration voltages, weights, etc.), to solve different classification problems. We show that ideal nanomaterial choice depends upon problem complexity, with more complex problems being favoured by complex voltage dependence of conductivity and vice versa. Furthermore, we highlight how intrinsic nanomaterial electrical properties can be exploited by differing configuration parameters, clarifying the role and limitations of these techniques. These findings provide guidance for the rational design of nanomaterials and algorithms for future Evolution-in-Materio processors.

Electrical behaviour and evolutionary computation in thin films of bovine brain microtubules.

We report on the electrical behaviour of thin films of bovine brain microtubules (MTs). For samples in both their dried and hydrated states, the measured currents reveal a power law dependence on the applied DC voltage. We attribute this to the injection of space-charge from the metallic electrode(s). The MTs are thought to form a complex electrical network, which can be manipulated with an applied voltage. This feature has been exploited to undertake some experiments on the use of the MT mesh as a medium for computation. We show that it is possible to evolve MT films into binary classifiers following an evolution in materio approach. The accuracy of the system is, on average, similar to that of early carbon nanotube classifiers developed using the same methodology.