On the Cryptographic Hardness of Finding a Nash Equilibrium
summary
Shows that, assuming indistinguishability obfuscation with subexponential security plus one-way functions, End-of-the-Line (hence PPAD and Nash equilibrium) is sub-exponentially hard on average. Construction embeds a long path using a pseudo-random function with obfuscated successor/predecessor, proves indistinguishability via sparse-trigger techniques and puncturing, and resolves the predecessor issue by simulating it through reversible simulation (Bennett) and a verifiability trick. Concrete hardness parameters and an explicit hard distribution are given.
AI-written summary · about
chapters (13)
- 0:00Intro
- 0:16The Story Line
- 0:40Game Theory and Nash Equilibrium
- 1:43The CS Perspective
- 3:55The Class PPAD [Papadimitriou 94]
- 4:49End of the Line Problem (EOL)
- 6:11PPAD and NASH [Papadimitriou 94]
- 8:53Program Obfuscation
- 11:47Main Theorem
- 12:34Hard EOL Instances
- 17:57A Useful Lemma
- 19:07Step 2 - Proof
- 21:09The Problem of Going Backwards
related talks
59mBeyond P versus NP: Using Complexity Theory in Obfuscation and Vice-Versa
58mObfuscation 2
01h 08mExpedition to Obfustopia: Indistinguishability Obfuscation from Well-Studied Assumptions to New...
50mApplications of Obfuscation II
48mTales of Obfuscation in Bounded Arithmetic, Metacomplexity, and Differential Privacy
01h 26mObfuscation 1
46mOn Hardness Assumptions Needed for ``Extreme High-End'' PRGs and Fast Derandomization
31mIndistinguishability Obfuscation, Range Avoidance, and Bounded Arithmetic
