Jingge Zhu

We consider the adaptive control problem for discrete-time, nonlinear stochastic systems with linearly parameterised uncertainty. Assuming access to a parameterised family of controllers that can stabilise the system in a bounded set within an informative region of the state space when the parameter is well-chosen, we propose a certainty equivalence learning-based adaptive control strategy, and subsequently derive stability bounds on the closed-loop system that hold for some probabilities. We then show that if the entire state space is informative, and the family of controllers is globally stabilising with appropriately chosen parameters, high probability stability guarantees can be derived.

Compute-Forward Multiple Access (CFMA) is a multiple access transmission scheme based on Compute-and-Forward (CF), which allows the receiver to first decode linear combinations of the transmitted signals and then solve for individual messages. This paper extends the CFMA scheme to a two-user Gaussian multiple-input multiple-output (MIMO) multiple access channel (MAC). We propose the CFMA Serial Coding Scheme (SCS) and the CFMA Parallel Coding Scheme (PCS) with nested lattice codes. We first derive the expression of the achievable rate pair for MIMO MAC with CFMA-SCS. We prove a general condition under which CFMA-SCS can achieve the sum capacity of the channel. Furthermore, this result is specialized to single-input multiple-output (SIMO) and 2-by-2 diagonal MIMO multiple access channels, for which more explicit sum capacity-achieving conditions on power and channel matrices are derived. We then study the achievable rate of CFMA-PCS by using an equivalent SIMO model, and analyze its sum capacity-achieving conditions. Numerical results are provided for the performance of CFMA-SCS and CFMA-PCS in different channel conditions. In general, CFMA-PCS has better sum capacity achievability, although with a higher computational complexity for encoding and decoding.

The generalization error of a learning algorithm refers to the discrepancy between the loss of a learning algorithm on training data and that on unseen testing data. Various information-theoretic bounds on the generalization error have been derived in the literature, where the mutual information between the training data and the hypothesis (the output of the learning algorithm) plays an important role. Focusing on the individual sample mutual information bound by (Bu et al., 2020), which itself is a tightened version of the first bound on the topic by (Russo and Zou, 2016) and (Xu and Raginsky, 2017), this paper investigates the tightness of these bounds, in terms of the dependence of their convergence rates on the sample size n. It has been recognized that these bounds are in general not tight, readily verified for the exemplary quadratic Gaussian mean estimation problem, where the individual sample mutual information bound scales as $O(\sqrt {1/n})$ while the true generalization error scales as $O(1/n)$ . The first contribution of this paper is to show that the same bound can in fact be asymptotically tight if an appropriate assumption is made. In particular, we show that the fast rate can be recovered when the assumption is made on the excess risk instead of the loss function, which was usually done in existing literature. A theoretical justification is given for this choice. The second contribution of the paper is a new set of generalization error bounds based on the $(\eta, c)$ -central condition, a condition relatively easy to verify and has the property that the mutual information term directly determines the convergence rate of the bound. Several analytical and numerical examples are given to show the effectiveness of these bounds.

As the number of mobile devices continues to grow, interference has become a major bottleneck in improving data rates in wireless networks. Efficient joint channel and power allocation (JCPA) is crucial for managing interference. In this paper, we first propose an enhanced WMMSE (eWMMSE) algorithm to solve the JCPA problem in multi-channel wireless networks. To reduce the computational complexity of iterative optimization, we further introduce JCPGNN-M, a graph neural network-based solution that enables simultaneous multi-channel allocation for each user. We reformulate the problem as a Lagrangian function, which allows us to enforce the total power constraints systematically. Our solution involves combining this Lagrangian framework with GNNs and iteratively updating the Lagrange multipliers and resource allocation scheme. Unlike existing GNN-based methods that limit each user to a single channel, JCPGNN-M supports efficient spectrum reuse and scales well in dense network scenarios. Simulation results show that JCPGNN-M achieves better data rate compared to eWMMSE. Meanwhile, the inference time of JCPGNN-M is much lower than eWMMS, and it can generalize well to larger networks.

Compute-forward multiple access (CFMA) is a transmission strategy which allows the receiver in a multiple access channel (MAC) to first decode linear combinations of the transmitted signals and then solve for individual messages. Compared to existing MAC strategies such as joint decoding or successive interference cancellation (SIC), CFMA was shown to achieve the MAC capacity region for fixed channels under certain signal-to-noise (SNR) conditions without time-sharing using only single-user decoders. This paper studies the CFMA scheme for a two-user Gaussian fast fading MAC with channel state information only available at the receiver (CSIR). We investigate appropriate lattice decoding schemes to decode linear combinations with any integer coefficients in the fading MAC and derive the achievable rate pairs. We give a sufficient and necessary condition under which the proposed scheme can achieve the ergodic sum capacity. Furthermore, we investigate the impact of channel statistics on the capacity achievability of the CFMA scheme. In general, the sum capacity is achievable if the channel variance is small compared to the mean value of the channel strengths. Various numerical results are presented to illustrate the theoretical findings.

Neural networks often have identifiable computational structures - components of the network which perform an interpretable algorithm or task - but the mechanisms by which these emerge and the best methods for detecting these structures are not well understood. In this paper we investigate the emergence of computational structure in a transformer-like model trained to simulate the physics of a particle system, where the transformer's attention mechanism is used to transfer information between particles. We show that (a) structures emerge in the attention heads of the transformer which learn to detect particle collisions, (b) the emergence of these structures is associated to degenerate geometry in the loss landscape, and (c) the dynamics of this emergence follows a power law. This suggests that these components are governed by a degenerate "effective potential". These results have implications for the convergence time of computational structure within neural networks and suggest that the emergence of computational structure can be detected by studying the dynamics of network components.

We study communication over a Gaussian multiple-access channel (MAC) with two types of transmitters: Digital transmitters hold a message from a discrete set that needs to be communicated to the receiver with vanishing error probability. Analog transmitters hold sequences of analog values. Some functions of these distributed values (but not the values themselves) need to be conveyed to the receiver, subject to a fidelity criterion such as mean squared error (MSE) or a certain maximum error with given confidence. For the case in which the computed function for the analog transmitters is a sum of values in $[-1,1]$ , we derive inner and outer bounds for the tradeoff of digital and analog rates of communication under peak and average power constraints for digital transmitters and a peak power constraint for analog transmitters. We then extend the achievability result to a class of functions that includes all linear and some non-linear functions. This extended scheme works over fading channels as long as full channel state information is available at the transmitter. The practicality of our proposed communication scheme is shown in channel simulations that use a version of the scheme based on low density parity check (LDPC) coding. We evaluate the system performance for different block lengths and Gaussian as well as non-Gaussian noise distributions.

We consider the problem of least squares parameter estimation from single-trajectory data for discrete-time, unstable, closed-loop nonlinear stochastic systems, with linearly parameterised uncertainty. Assuming a region of the state space produces informative data, and the system is sub-exponentially unstable, we establish non-asymptotic guarantees on the estimation error at times where the state trajectory evolves in this region. If the whole state space is informative, high probability guarantees on the error hold for all times. Examples are provided where our results are useful for analysis, but existing results are not.

Transfer learning, or domain adaptation, is concerned with machine learning problems in which training and testing data come from possibly different probability distributions. In this work, we give an information-theoretic analysis of the generalization error and excess risk of transfer learning algorithms. Our results suggest, perhaps as expected, that the Kullback-Leibler (KL) divergence D (μ∥μ′) plays an important role in the characterizations where μ and μ′ denote the distribution of the training data and the testing data, respectively. Specifically, we provide generalization error and excess risk upper bounds for learning algorithms where data from both distributions are available in the training phase. Recognizing that the bounds could be sub-optimal in general, we provide improved excess risk upper bounds for a certain class of algorithms, including the empirical risk minimization (ERM) algorithm, by making stronger assumptions through the central condition . To demonstrate the usefulness of the bounds, we further extend the analysis to the Gibbs algorithm and the noisy stochastic gradient descent method. We then generalize the mutual information bound with other divergences such as ϕ-divergence and Wasserstein distance, which may lead to tighter bounds and can handle the case when μ is not absolutely continuous with respect to μ′. Several numerical results are provided to demonstrate our theoretical findings. Lastly, to address the problem that the bounds are often not directly applicable in practice due to the absence of the distributional knowledge of the data, we develop an algorithm (called InfoBoost) that dynamically adjusts the importance weights for both source and target data based on certain information measures. The empirical results show the effectiveness of the proposed algorithm.