Tilahun M. Getu

Tilahun M. Getu, Walid Saad, Georges Kaddoum et al.

Although deep learning (DL)-enabled semantic communication (SemCom) has emerged as a 6G enabler by minimizing irrelevant information transmission -- minimizing power usage, bandwidth consumption, and transmission delay, its benefits can be limited by radio frequency interference (RFI) that causes substantial semantic noise. Such semantic noise's impact can be alleviated using an interference-resistant and robust (IR$^2$) SemCom design, though no such design exists yet. To stimulate fundamental research on IR2 SemCom, the performance limits of a popular text SemCom system named DeepSC are studied in the presence of (multi-interferer) RFI. By introducing a principled probabilistic framework for SemCom, we show that DeepSC produces semantically irrelevant sentences as the power of (multi-interferer) RFI gets very large. We also derive DeepSC's practical limits and a lower bound on its outage probability under multi-interferer RFI, and propose a (generic) lifelong DL-based IR$^2$ SemCom system. We corroborate the derived limits with simulations and computer experiments, which also affirm the vulnerability of DeepSC to a wireless attack using RFI.

Tilahun M. Getu, Nada T. Golmie, David W. Griffith

We provide a new generation solution to the fundamental old problem of a doubly selective fading channel estimation for orthogonal frequency division multiplexing (OFDM) systems. For systems based on OFDM, we propose a deep learning (DL)-based blind doubly selective channel estimator. This estimator does require no pilot symbols, unlike the corresponding state-of-the-art estimators, even during the estimation of a deep fading doubly selective channel. We also provide the first of its kind theory on the testing mean squared error (MSE) performance of our investigated blind OFDM channel estimator based on over-parameterized ReLU FNNs.

Tilahun M. Getu, Georges Kaddoum, M. Bennis

Although deep learning (DL) has led to several breakthroughs in many disciplines, the fundamental understanding on why and how DL is empirically successful remains elusive. To attack this fundamental problem and unravel the mysteries behind DL's empirical successes, significant innovations toward a unified theory of DL have been made. Although these innovations encompass nearly fundamental advances in optimization, generalization, and approximation, no work has quantified the testing performance of a DL-based algorithm employed to solve a pattern classification problem. To overcome this fundamental challenge in part, this paper exposes the fundamental testing performance limits of DL-based binary classifiers trained with hinge loss. For binary classifiers that are based on deep rectified linear unit (ReLU) feedforward neural networks (FNNs) and deep FNNs with ReLU and Tanh activation, we derive their respective novel asymptotic testing performance limits, which are validated by extensive computer experiments.

Tilahun M. Getu

Among the several paradigms of artificial intelligence (AI) or machine learning (ML), a remarkably successful paradigm is deep learning. Deep learning's phenomenal success has been hoped to be interpreted via fundamental research on the theory of deep learning. Accordingly, applied research on deep learning has spurred the theory of deep learning-oriented depth and breadth of developments. Inspired by such developments, we pose these fundamental questions: can we accurately approximate an arbitrary matrix-vector product using deep rectified linear unit (ReLU) feedforward neural networks (FNNs)? If so, can we bound the resulting approximation error? In light of these questions, we derive error bounds in Lebesgue and Sobolev norms that comprise our developed deep approximation theory. Guided by this theory, we have successfully trained deep ReLU FNNs whose test results justify our developed theory. The developed theory is also applicable for guiding and easing the training of teacher deep ReLU FNNs in view of the emerging teacher-student AI or ML paradigms that are essential for solving several AI or ML problems in wireless communications and signal processing; network science and graph signal processing; and network neuroscience and brain physics.