Depth-Width Trade-offs for ReLU Networks via Sharkovsky's Theorem

Abstract

Understanding the representational power of Deep Neural Networks (DNNs) and how their structural properties (e.g., depth, width, type of activation unit) affect the functions they can compute, has been an important yet challenging question in deep learning and approximation theory. In a seminal paper, Telgarsky highlighted the benefits of depth by presenting a family of functions (based on simple triangular waves) for which DNNs achieve zero classification error, whereas shallow networks with fewer than exponentially many nodes incur constant error. Even though Telgarsky's work reveals the limitations of shallow neural networks, it does not inform us on why these functions are difficult to represent and in fact he states it as a tantalizing open question to characterize those functions that cannot be well-approximated by smaller depths. In this work, we point to a new connection between DNNs expressivity and Sharkovsky's Theorem from dynamical systems, that enables us to characterize the depth-width trade-offs of ReLU networks for representing functions based on the presence of generalized notion of fixed points, called periodic points (a fixed point is a point of period 1). Motivated by our observation that the triangle waves used in Telgarsky's work contain points of period 3 - a period that is special in that it implies chaotic behavior based on the celebrated result by Li-Yorke - we proceed to give general lower bounds for the width needed to represent periodic functions as a function of the depth. Technically, the crux of our approach is based on an eigenvalue analysis of the dynamical system associated with such functions.

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Representation Benefits of Deep Feedforward Networks
• Understanding Deep Neural Networks with Rectified Linear Units
• Newton vs the machine: solving the chaotic three-body problem using deep neural networks
• On the Expressive Efficiency of Sum Product Networks
• On the Expressive Power of Deep Polynomial Neural Networks
• Is Deeper Better only when Shallow is Good?

Cited by in corpus (9)

• Depth-Width Trade-offs for Neural Networks via Topological Entropy
• Statistically Meaningful Approximation: a Case Study on Approximating Turing Machines with Transformers
• Sharp Representation Theorems for ReLU Networks with Precise Dependence on Depth
• The Connection Between Approximation, Depth Separation and Learnability in Neural Networks
• On the Optimal Memorization Power of ReLU Neural Networks
• Depth separation beyond radial functions
• On the Number of Linear Functions Composing Deep Neural Network: Towards a Refined Definition of Neural Networks Complexity
• PDE constraints on smooth hierarchical functions computed by neural networks
• Expressivity of Neural Networks via Chaotic Itineraries beyond Sharkovsky's Theorem