Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
Related publications (38)
Graph Chatbot
Chat with Graph Search
Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.
DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.
This paper presents a systematic computational study on the performance of distributed optimization in model predictive control (MPC). We consider networks of dynamically coupled systems, which are subject to input and state con- straints. The resulting MP ...
This work characterizes the nature of the limit point of distributed strategies for adaptation and learning over networks in the general case when the combination policy is not necessarily doubly stochastic and when the individual risks do not necessarily ...
Recent experimental analysis has shown that some types of urban networks exhibit a low scatter reproducible relationship between average network flow and density, known as the macroscopic fundamental diagram (MFD). It has also been shown that heterogeneity ...
Adaptive networks rely on in-network and collaborative processing among distributed agents to deliver enhanced performance in estimation and inference tasks. Information is exchanged among the nodes, usually over noisy links. The combination weights that a ...
Observations of traffic pairs of flow vs. density or occupancy for individual locations in freeways or arterials are usually scattered about an underlying curve. Recent observations from empirical data in arterial networks showed that in some cases by aggr ...
Complex networks exist for a number of purposes. The neural, metabolic and food networks ensure our survival, while the social, economic, transportation and communication networks allow us to prosper. Independently of the purposes and particularities of th ...
Ensuring correct network behavior is hard. This is the case even for simple networks, and adding middleboxes only complicates this task. In this paper, we demonstrate a fundamental property of networks. Namely, we show a way of using a network to emulate t ...
Observations of traffic pairs of flow vs. density or occupancy for individual locations in freeways or arterials are usually scattered about an underlying curve. Recent observations from empirical data in arterial networks showed that in some cases by aggr ...
Elsevier Science, Reg Sales Off, Customer Support Dept, 655 Ave Of The Americas, New York, Ny 10010 Usa2011
How can we localize the source of diffusion in a complex network? Due to the tremendous size of many real networks---such as the Internet or the human social graph---it is usually infeasible to observe the state of all nodes in a network. We show that it i ...
The depolarization temperature T-d of piezoelectric materials is an important figure of merit for their application at elevated temperatures. Until now, there are several methods proposed in the literature to determine the depolarization temperature of pie ...