Source author record

Lei M. Zhang

Lei M. Zhang appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Information Theory math.IT Artificial Intelligence Machine Learning Multiagent Systems

Catalog footprint

What is connected

3works

5topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

Learning Robust Real-Time Cultural Transmission without Human Data

Cultural transmission is the domain-general social skill that allows agents to acquire and use information from each other in real-time with high fidelity and recall. In humans, it is the inheritance process that powers cumulative cultural evolution, expanding our skills, tools and knowledge across generations. We provide a method for generating zero-shot, high recall cultural transmission in artificially intelligent agents. Our agents succeed at real-time cultural transmission from humans in novel contexts without using any pre-collected human data. We identify a surprisingly simple set of ingredients sufficient for generating cultural transmission and develop an evaluation methodology for rigorously assessing it. This paves the way for cultural evolution as an algorithm for developing artificial general intelligence.

preprint2016arXiv

Feed-Forward Staircase Codes

We propose two variants of staircase codes that resolve the issue of parity-propagation in their encoding process. The proposed codes provide a systematic way of terminating a staircase code after an arbitrary number of blocks. The class of feed-forward staircase codes are introduced, which uses a self-protection technique to avoid parity-propagation. We also introduce the class of partial feed-forward staircase codes, which allows parity-propagation to occur over a given number of blocks. By amortizing the complexity of self-protection over several standard staircase blocks, the encoding complexity of these codes is made comparable to staircase codes. Partial feed-forward staircase codes have the same error-floor as staircase codes. Simulations confirm that the performance of the proposed codes in both the waterfall and error-floor regions is similar to the original staircase codes. The proposed codes help extend the domain of application of staircase codes to systems in which parity-propagation is undesirable or termination is necessary.

preprint2015arXiv

Spatially-coupled Split-component Codes with Iterative Algebraic Decoding

We analyze a class of high performance, low decoding-data-flow error-correcting codes suitable for high bit-rate optical-fiber communication systems. A spatially-coupled split-component ensemble is defined, generalizing from the most important codes of this class, staircase codes and braided block codes, and preserving a deterministic partitioning of component-code bits over code blocks. Our analysis focuses on low-complexity iterative algebraic decoding, which, for the binary erasure channel, is equivalent to a generalization of the peeling decoder. Using the differential equation method, we derive a vector recursion that tracks the expected residual graph evolution throughout the decoding process. The threshold of the recursion is found using potential function analysis. We generalize the analysis to mixture ensembles consisting of more than one type of component code, which provide increased flexibility of ensemble parameters and can improve performance. The analysis extends to the binary symmetric channel by assuming mis-correction-free component-code decoding. Simple upper-bounds on the number of errors correctable by the ensemble are derived. Finally, we analyze the threshold of spatially-coupled split-component ensembles under beyond bounded-distance component decoding.