Arora. Overcoming Intractability in Machine Learning
A seminar course that will focus on the following phenomenon: many problems in machine learning are formally intractable (e.g., NP-hard). Nevertheless they are solved in practice by heuristics. Can we design algorithms with provable guarantees (running time, solution quality)?
- from
- Theoretical Computer Science
- added
- 2026-10-10
- likes
- 0
similar
-
Arora. Advanced Algorithm Design cs.princeton.edu
Notably uses ideas such as randomness, approximation, high dimensional geometry. Faces uncertainty, approaches to handle big data, handling intractability, heuristic approaches, ..etc.
-
Arora. A Theorist's Toolkit. Princeton cs.princeton.edu
Aimed primarily at first and second year graduate students who plan to do research in theoretical computer science. We will introduce probabilistic, algebraic, combinatorial, and algorithmic methods useful in proofs.
-
Arora. Thinking Like a Theorist. Princeton cs.princeton.edu
It covers a large number of the math/CS topics that you need to know for reading and doing research in Computer Science Theory.
-
Demaine, Gasarch & Hajiaghayi. Computers and Intractability: A Guide to Algorithmic Lower Bounds hardness.mit.edu
A sequel to Garey and Johnson's Computers and Intractability: A Guide to NP-Completeness. New topics include Parameterized Complexity, Lower bounds on approximation, Other hardness assumptions (ETH, 3SUM-conjecture, APSP-conjecture, UGC, Others), Online Algorithms, Streaming…
-
Efficient Algorithms and Intractable Problems, 2001 people.eecs.berkeley.edu
-
Arora. The Computational Universe cs.princeton.edu
Takes us on a broad sweep of scientific knowledge and related technologies: propositional logic of the ancient Greeks (microprocessors); quantum mechanics (silicon chips); network and system phenomena (internet and search engines); computational intractability (secure…
Theoretical Computer Science › Lecture Notes: “A seminar course that will focus on the following phenomenon: many problems in machine learning are formally intractable (e.g., NP-hard). Nevertheless they are solved in practice by heuristics. Can we design algorithms with provable guarantees (running time, solution quality)?”