Consensus Through Herding
summary
Presents a protocol that achieves communication-efficient adaptive security for consensus without proof-of-work or erasure assumptions, using standard crypto (sub‑exponential hardness) in the synchronous diffusion model. Key tools are a new batch agreement from herding construction, transaction‑specific VRF eligibility (VRF per (round,transaction)) plus an exponential‑decay initial score for blocks, and probability bounds (Chernoff/union) to block grinding. Guarantees: polylogarithmic communication and confirmation time in the number of nodes, tolerates adaptive corruptions, and the constructed protocol is proven to satisfy consistency and strong liveness.
AI-written summary · about
chapters (13)
- 0:00Personal introduction
- 1:54Consensus model definition
- 7:53Consensus implementation
- 11:49Byzantine agreement basics
- 13:28Scalability and security
- 17:45Adaptive security challenges
- 20:44Existing solutions overview
- 22:15VRF and leader election
- 27:25The need for new solutions
- 31:03Introduction to herding
- 34:24Defining batch agreement
- 38:15Economic theory of herding
- 46:42Protocol and security proof
related talks
42mBeyond Blockchain: Convergent Consensus
47mCant we just agree
16mDispersedLedger: High-Throughput Byzantine Consensus on Variable Bandwidth Networks
56mA sledgehammer to crack a nut: why blockchain is not (always) a good idea
19mThe Time Complexity of Consensus Under Oblivious Message Adversaries
40mConsensus Algorithms in Distributed Systems
22mOptimizing network topologies...
14mBolt-On Strong Consistency: Specification, Implementation, and Verification