The problem seems to have been stated in the mid-20th century after work by Beurling and von Neumann,1 who found (but never published) a positive solution for the case of compact operators. It was then posed by Paul Halmos for the case of operators T {\displaystyle T} such that T 2 {\displaystyle T^{2}} is compact. This was resolved affirmatively, for the more general class of polynomially compact operators (operators T {\displaystyle T} such that p ( T ) {\displaystyle p(T)} is a compact operator for a suitably chosen non-zero polynomial p {\displaystyle p} ), by Allen R. Bernstein and Abraham Robinson in 1966 (see Non-standard analysis § Invariant subspace problem for a summary of the proof).
For Banach spaces, the first example of an operator without an invariant subspace was constructed by Per Enflo. He proposed a counterexample to the invariant subspace problem in 1975, publishing an outline in 1976. Enflo submitted the full article in 1981 and the article's complexity and length delayed its publication to 19872 Enflo's long "manuscript had a world-wide circulation among mathematicians"3 and some of its ideas were described in publications besides Enflo (1976).4 Enflo's works inspired a similar construction of an operator without an invariant subspace for example by Bernard Beauzamy, who acknowledged Enflo's ideas.5
In the 1990s, Enflo developed a "constructive" approach to the invariant subspace problem on Hilbert spaces.6
In May 2023, a preprint of Enflo appeared on arXiv,7 which, if correct, solves the problem for Hilbert spaces and completes the picture.
In July 2023, a second and independent preprint of Neville appeared on arXiv,8 claiming the solution of the problem for separable Hilbert spaces.
In September 2024, a peer-reviewed article published in Axioms by a team of four Jordanian academic researchers announced that they had solved the invariant subspace problem.9 However, basic mistakes in the proof were pointed out.1011
Formally, the invariant subspace problem for a complex Banach space H {\displaystyle H} of dimension > 1 is the question whether every bounded linear operator T : H → H {\displaystyle T:H\to H} has a non-trivial closed T {\displaystyle T} -invariant subspace: a closed linear subspace W {\displaystyle W} of H {\displaystyle H} , which is different from { 0 } {\displaystyle \{0\}} and from H {\displaystyle H} , such that T ( W ) ⊂ W {\displaystyle T(W)\subset W} .
A negative answer to the problem is closely related to properties of the orbits T {\displaystyle T} . If x {\displaystyle x} is an element of the Banach space H {\displaystyle H} , the orbit of x {\displaystyle x} under the action of T {\displaystyle T} , denoted by [ x ] {\displaystyle [x]} , is the subspace generated by the sequence { T n ( x ) : n ≥ 0 } {\displaystyle \{T^{n}(x)\,:\,n\geq 0\}} . This is also called the T {\displaystyle T} -cyclic subspace generated by x {\displaystyle x} . From the definition it follows that [ x ] {\displaystyle [x]} is a T {\displaystyle T} -invariant subspace. Moreover, it is the minimal T {\displaystyle T} -invariant subspace containing x {\displaystyle x} : if W {\displaystyle W} is another invariant subspace containing x {\displaystyle x} , then necessarily T n ( x ) ∈ W {\displaystyle T^{n}(x)\in W} for all n ≥ 0 {\displaystyle n\geq 0} (since W {\displaystyle W} is T {\displaystyle T} -invariant), and so [ x ] ⊂ W {\displaystyle [x]\subset W} . If x {\displaystyle x} is non-zero, then [ x ] {\displaystyle [x]} is not equal to { 0 } {\displaystyle \{0\}} , so its closure is either the whole space H {\displaystyle H} (in which case x {\displaystyle x} is said to be a cyclic vector for T {\displaystyle T} ) or it is a non-trivial T {\displaystyle T} -invariant subspace. Therefore, a counterexample to the invariant subspace problem would be a Banach space H {\displaystyle H} and a bounded operator T : H → H {\displaystyle T:H\to H} for which every non-zero vector x ∈ H {\displaystyle x\in H} is a cyclic vector for T {\displaystyle T} . (Where a "cyclic vector" x {\displaystyle x} for an operator T {\displaystyle T} on a Banach space H {\displaystyle H} means one for which the orbit [ x ] {\displaystyle [x]} of x {\displaystyle x} is dense in H {\displaystyle H} .)
While the case of the invariant subspace problem for separable Hilbert spaces is still open, several other cases have been settled for topological vector spaces (over the field of complex numbers):
Yadav (2005), p. 292. - Yadav, B. S. (2005), "The present state and heritages of the invariant subspace problem", Milan Journal of Mathematics, 73 (1): 289–316, doi:10.1007/s00032-005-0048-7, MR 2175046, S2CID 121068326 https://doi.org/10.1007%2Fs00032-005-0048-7 ↩
Beauzamy (1988); Yadav (2005). - Beauzamy, Bernard (1988), Introduction to operator theory and invariant subspaces, North-Holland Mathematical Library, vol. 42, Amsterdam: North-Holland, ISBN 978-0-444-70521-1, MR 0967989 https://mathscinet.ams.org/mathscinet-getitem?mr=0967989 ↩
See, for example, Radjavi & Rosenthal (1982). - Radjavi, Heydar; Rosenthal, Peter (1982), "The invariant subspace problem", The Mathematical Intelligencer, 4 (1): 33–37, doi:10.1007/BF03022994, MR 0678734, S2CID 122811130 https://link.springer.com/article/10.1007/BF03022994 ↩
Page 401 in Foiaş, Ciprian; Jung, Il Bong; Ko, Eungil; Pearcy, Carl (2005). "On quasinilpotent operators. III". Journal of Operator Theory. 54 (2): 401–414.. Enflo's method of ("forward") "minimal vectors" is also noted in the review of this research article by Gilles Cassier in Mathematical Reviews: MR2186363 /wiki/Mathematical_Reviews ↩
Enflo, Per H. (May 26, 2023). "On the invariant subspace problem in Hilbert spaces". arXiv:2305.15442 [math.FA]. /wiki/ArXiv_(identifier) ↩
Neville, Charles W. (July 21, 2023). "a proof of the invariant subspace conjecture for separable Hilbert spaces". arXiv:2307.08176 [math.FA]. /wiki/ArXiv_(identifier) ↩
Khalil, Roshdi; Yousef, Abdelrahman; Alshanti, Waseem Ghazi; Hammad, Ma’mon Abu (2024-09-02). "The Invariant Subspace Problem for Separable Hilbert Spaces". Axioms. 13 (9): 598. doi:10.3390/axioms13090598. ISSN 2075-1680. https://doi.org/10.3390%2Faxioms13090598 ↩
Ghatasheh, Ahmed (Nov 28, 2024). "Refuting a recent proof of the invariant subspace problem". arXiv:2411.19409 [math.FA]. /wiki/ArXiv_(identifier) ↩
See mathoverflow: [1]. https://mathoverflow.net/questions/478156/a-claim-for-a-proof-of-the-invariant-subspace-problem ↩
Von Neumann's proof was never published, as relayed in a private communication to the authors of Aronszajn & Smith (1954). A version of that proof, independently discovered by Aronszajn, is included at the end of that paper. - Aronszajn, N.; Smith, K. T. (1954), "Invariant subspaces of completely continuous operators", Annals of Mathematics, Second Series, 60 (2): 345–350, doi:10.2307/1969637, JSTOR 1969637, MR 0065807 https://doi.org/10.2307%2F1969637 ↩
See Pearcy & Shields (1974) for a review. - Pearcy, Carl; Shields, Allen L. (1974), "A survey of the Lomonosov technique in the theory of invariant subspaces", in C. Pearcy (ed.), Topics in operator theory, Mathematical Surveys, Providence, R.I.: American Mathematical Society, pp. 219–229, MR 0355639 https://mathscinet.ams.org/mathscinet-getitem?mr=0355639 ↩