Brendan Tracey

Brendan D. Tracey, Andrea Michi, Yuri Chervonyi et al. · deepmind

Reinforcement learning (RL) has shown promising results for real-time control systems, including the domain of plasma magnetic control. However, there are still significant drawbacks compared to traditional feedback control approaches for magnetic confinement. In this work, we address key drawbacks of the RL method; achieving higher control accuracy for desired plasma properties, reducing the steady-state error, and decreasing the required time to learn new tasks. We build on top of \cite{degrave2022magnetic}, and present algorithmic improvements to the agent architecture and training procedure. We present simulation results that show up to 65\% improvement in shape accuracy, achieve substantial reduction in the long-term bias of the plasma current, and additionally reduce the training time required to learn new tasks by a factor of 3 or more. We present new experiments using the upgraded RL-based controllers on the TCV tokamak, which validate the simulation results achieved, and point the way towards routinely achieving accurate discharges using the RL approach.

Brendan Tracey, David Wolpert, Juan J. Alonso

Monte Carlo (MC) techniques are often used to estimate integrals of a multivariate function using randomly generated samples of the function. In light of the increasing interest in uncertainty quantification and robust design applications in aerospace engineering, the calculation of expected values of such functions (e.g. performance measures) becomes important. However, MC techniques often suffer from high variance and slow convergence as the number of samples increases. In this paper we present Stacked Monte Carlo (StackMC), a new method for post-processing an existing set of MC samples to improve the associated integral estimate. StackMC is based on the supervised learning techniques of fitting functions and cross validation. It should reduce the variance of any type of Monte Carlo integral estimate (simple sampling, importance sampling, quasi-Monte Carlo, MCMC, etc.) without adding bias. We report on an extensive set of experiments confirming that the StackMC estimate of an integral is more accurate than both the associated unprocessed Monte Carlo estimate and an estimate based on a functional fit to the MC samples. These experiments run over a wide variety of integration spaces, numbers of sample points, dimensions, and fitting functions. In particular, we apply StackMC in estimating the expected value of the fuel burn metric of future commercial aircraft and in estimating sonic boom loudness measures. We compare the efficiency of StackMC with that of more standard methods and show that for negligible additional computational cost significant increases in accuracy are gained.

Jonas Buchli, Brendan Tracey, Tomislav Andric et al. · deepmind

Improved low-frequency sensitivity of gravitational wave observatories would unlock study of intermediate-mass black hole mergers, binary black hole eccentricity, and provide early warnings for multi-messenger observations of binary neutron star mergers. Today's mirror stabilization control injects harmful noise, constituting a major obstacle to sensitivity improvements. We eliminated this noise through Deep Loop Shaping, a reinforcement learning method using frequency domain rewards. We proved our methodology on the LIGO Livingston Observatory (LLO). Our controller reduced control noise in the 10--30Hz band by over 30x, and up to 100x in sub-bands surpassing the design goal motivated by the quantum limit. These results highlight the potential of Deep Loop Shaping to improve current and future GW observatories, and more broadly instrumentation and control systems.

Siqi Liu, Guy Lever, Zhe Wang et al.

Intelligent behaviour in the physical world exhibits structure at multiple spatial and temporal scales. Although movements are ultimately executed at the level of instantaneous muscle tensions or joint torques, they must be selected to serve goals defined on much longer timescales, and in terms of relations that extend far beyond the body itself, ultimately involving coordination with other agents. Recent research in artificial intelligence has shown the promise of learning-based approaches to the respective problems of complex movement, longer-term planning and multi-agent coordination. However, there is limited research aimed at their integration. We study this problem by training teams of physically simulated humanoid avatars to play football in a realistic virtual environment. We develop a method that combines imitation learning, single- and multi-agent reinforcement learning and population-based training, and makes use of transferable representations of behaviour for decision making at different levels of abstraction. In a sequence of stages, players first learn to control a fully articulated body to perform realistic, human-like movements such as running and turning; they then acquire mid-level football skills such as dribbling and shooting; finally, they develop awareness of others and play as a team, bridging the gap between low-level motor control at a timescale of milliseconds, and coordinated goal-directed behaviour as a team at the timescale of tens of seconds. We investigate the emergence of behaviours at different levels of abstraction, as well as the representations that underlie these behaviours using several analysis techniques, including statistics from real-world sports analytics. Our work constitutes a complete demonstration of integrated decision-making at multiple scales in a physically embodied multi-agent setting. See project video at https://youtu.be/KHMwq9pv7mg.

Wojciech Marian Czarnecki, Gauthier Gidel, Brendan Tracey et al.

This paper investigates the geometrical properties of real world games (e.g. Tic-Tac-Toe, Go, StarCraft II). We hypothesise that their geometrical structure resemble a spinning top, with the upright axis representing transitive strength, and the radial axis, which corresponds to the number of cycles that exist at a particular transitive strength, representing the non-transitive dimension. We prove the existence of this geometry for a wide class of real world games, exposing their temporal nature. Additionally, we show that this unique structure also has consequences for learning - it clarifies why populations of strategies are necessary for training of agents, and how population size relates to the structure of the game. Finally, we empirically validate these claims by using a selection of nine real world two-player zero-sum symmetric games, showing 1) the spinning top structure is revealed and can be easily re-constructed by using a new method of Nash clustering to measure the interaction between transitive and cyclical strategy behaviour, and 2) the effect that population size has on the convergence in these games.

Artemy Kolchinsky, Brendan D. Tracey, Steven Van Kuyk

Information bottleneck (IB) is a method for extracting information from one random variable $X$ that is relevant for predicting another random variable $Y$. To do so, IB identifies an intermediate "bottleneck" variable $T$ that has low mutual information $I(X;T)$ and high mutual information $I(Y;T)$. The "IB curve" characterizes the set of bottleneck variables that achieve maximal $I(Y;T)$ for a given $I(X;T)$, and is typically explored by maximizing the "IB Lagrangian", $I(Y;T) - βI(X;T)$. In some cases, $Y$ is a deterministic function of $X$, including many classification problems in supervised learning where the output class $Y$ is a deterministic function of the input $X$. We demonstrate three caveats when using IB in any situation where $Y$ is a deterministic function of $X$: (1) the IB curve cannot be recovered by maximizing the IB Lagrangian for different values of $β$; (2) there are "uninteresting" trivial solutions at all points of the IB curve; and (3) for multi-layer classifiers that achieve low prediction error, different layers cannot exhibit a strict trade-off between compression and prediction, contrary to a recent proposal. We also show that when $Y$ is a small perturbation away from being a deterministic function of $X$, these three caveats arise in an approximate way. To address problem (1), we propose a functional that, unlike the IB Lagrangian, can recover the IB curve in all cases. We demonstrate the three caveats on the MNIST dataset.

Kunal Menda, Yi-Chun Chen, Justin Grana et al.

The incorporation of macro-actions (temporally extended actions) into multi-agent decision problems has the potential to address the curse of dimensionality associated with such decision problems. Since macro-actions last for stochastic durations, multiple agents executing decentralized policies in cooperative environments must act asynchronously. We present an algorithm that modifies generalized advantage estimation for temporally extended actions, allowing a state-of-the-art policy optimization algorithm to optimize policies in Dec-POMDPs in which agents act asynchronously. We show that our algorithm is capable of learning optimal policies in two cooperative domains, one involving real-time bus holding control and one involving wildfire fighting with unmanned aircraft. Our algorithm works by framing problems as "event-driven decision processes," which are scenarios in which the sequence and timing of actions and events are random and governed by an underlying stochastic process. In addition to optimizing policies with continuous state and action spaces, our algorithm also facilitates the use of event-driven simulators, which do not require time to be discretized into time-steps. We demonstrate the benefit of using event-driven simulation in the context of multiple agents taking asynchronous actions. We show that fixed time-step simulation risks obfuscating the sequence in which closely separated events occur, adversely affecting the policies learned. In addition, we show that arbitrarily shrinking the time-step scales poorly with the number of agents.

Artemy Kolchinsky, Brendan D. Tracey, David H. Wolpert

Information bottleneck (IB) is a technique for extracting information in one random variable $X$ that is relevant for predicting another random variable $Y$. IB works by encoding $X$ in a compressed "bottleneck" random variable $M$ from which $Y$ can be accurately decoded. However, finding the optimal bottleneck variable involves a difficult optimization problem, which until recently has been considered for only two limited cases: discrete $X$ and $Y$ with small state spaces, and continuous $X$ and $Y$ with a Gaussian joint distribution (in which case optimal encoding and decoding maps are linear). We propose a method for performing IB on arbitrarily-distributed discrete and/or continuous $X$ and $Y$, while allowing for nonlinear encoding and decoding maps. Our approach relies on a novel non-parametric upper bound for mutual information. We describe how to implement our method using neural networks. We then show that it achieves better performance than the recently-proposed "variational IB" method on several real-world datasets.

Brendan D. Tracey, David H. Wolpert

Monte Carlo (MC) sampling algorithms are an extremely widely-used technique to estimate expectations of functions f(x), especially in high dimensions. Control variates are a very powerful technique to reduce the error of such estimates, but in their conventional form rely on having an accurate approximation of f, a priori. Stacked Monte Carlo (StackMC) is a recently introduced technique designed to overcome this limitation by fitting a control variate to the data samples themselves. Done naively, forming a control variate to the data would result in overfitting, typically worsening the MC algorithm's performance. StackMC uses in-sample / out-sample techniques to remove this overfitting. Crucially, it is a post-processing technique, requiring no additional samples, and can be applied to data generated by any MC estimator. Our preliminary experiments demonstrated that StackMC improved the estimates of expectations when it was used to post-process samples produces by a "simple sampling" MC estimator. Here we substantially extend this earlier work. We provide an in-depth analysis of the StackMC algorithm, which we use to construct an improved version of the original algorithm, with lower estimation error. We then perform experiments of StackMC on several additional kinds of MC estimators, demonstrating improved performance when the samples are generated via importance sampling, Latin-hypercube sampling and quasi-Monte Carlo sampling. We also show how to extend StackMC to combine multiple fitting functions, and how to apply it to discrete input spaces x.

Erik J. Schlicht, Ritchie Lee, David H. Wolpert et al.

Multi-fidelity methods combine inexpensive low-fidelity simulations with costly but highfidelity simulations to produce an accurate model of a system of interest at minimal cost. They have proven useful in modeling physical systems and have been applied to engineering problems such as wing-design optimization. During human-in-the-loop experimentation, it has become increasingly common to use online platforms, like Mechanical Turk, to run low-fidelity experiments to gather human performance data in an efficient manner. One concern with these experiments is that the results obtained from the online environment generalize poorly to the actual domain of interest. To address this limitation, we extend traditional multi-fidelity approaches to allow us to combine fewer data points from high-fidelity human-in-the-loop experiments with plentiful but less accurate data from low-fidelity experiments to produce accurate models of how humans interact. We present both model-based and model-free methods, and summarize the predictive performance of each method under dierent conditions.

Erik J. Schlicht, Ritchie Lee, David H. Wolpert et al.

Multi-fidelity methods combine inexpensive low-fidelity simulations with costly but high-fidelity simulations to produce an accurate model of a system of interest at minimal cost. They have proven useful in modeling physical systems and have been applied to engineering problems such as wing-design optimization. During human-in-the-loop experimentation, it has become increasingly common to use online platforms, like Mechanical Turk, to run low-fidelity experiments to gather human performance data in an efficient manner. One concern with these experiments is that the results obtained from the online environment generalize poorly to the actual domain of interest. To address this limitation, we extend traditional multi-fidelity approaches to allow us to combine fewer data points from high-fidelity human-in-the-loop experiments with plentiful but less accurate data from low-fidelity experiments to produce accurate models of how humans interact. We present both model-based and model-free methods, and summarize the predictive performance of each method under different conditions.