Abstract:Subsolutions and concavity play critical roles in classical solvability, especially a priori estimates, of fully nonlinear elliptic equations. Our first primary goal in this paper is to explore the possibility to weaken the concavity condition. The second is to clarify relations between weak notions of subsolution introduced by Székelyhidi and the author, respectively, in attempt to treat equations on closed manifolds. More precisely, we show that these weak notions of subsolutions are equivalent for equations defined on convex cones of type 1 in the sense defined by Caffarelli, Nirenberg and Spruck.
What problem does this paper attempt to address?
The main problems that this paper attempts to solve are related to subsolutions and concavity conditions in fully nonlinear elliptic equations. Specifically, the author Bo Guan explores two main objectives in the paper:
1. **The possibility of weakening the concavity condition**: Traditionally, when studying a priori estimates and classical solvability of fully nonlinear elliptic equations, the function \( f \) is usually required to satisfy the concavity condition. One of the author's goals is to explore whether this condition can be relaxed in some cases, that is, to assume that \( f \) is concave outside a certain bounded set instead of being concave throughout the entire domain.
2. **The relationship between different weak subsolution concepts**: The author also aims to clarify the relationship between different weak subsolution concepts introduced by Székelyhidi and himself respectively, especially when dealing with equations on closed manifolds. Specifically, the author proves that these weak subsolution concepts are equivalent on a certain class of convex cones.
### Specific problem description
#### Weakening the concavity condition
The author proposes that when deriving a priori estimates, it can be assumed that \( f \) is concave outside a certain bounded set \( B_R(0) \) instead of being concave throughout the entire domain \( \Gamma \). Technically, this means:
\[ f \text{ is concave on } \Gamma \setminus B_R(0) \]
#### The relationship between different weak subsolution concepts
The author considers fully nonlinear elliptic equations defined on closed manifolds and discusses two different weak subsolution concepts:
- **The condition at the tangent cone at infinity**: Introduced by Guan in [6] and used to derive second - order estimates on closed manifolds.
- **\(\mathcal{C}\)-subsolution**: Introduced by Székelyhidi in [17] and used to handle the complex - valued version of the equation on closed Hermitian manifolds.
The author proves that on the type - 1 cone \( \Gamma \), these two weak subsolution concepts are equivalent.
### Main results
- **Theorem 1.3**: Under the assumption that \( f \) is concave on \( \Gamma \setminus B_R(0) \), the author proves that the a priori estimates of the gradient and the second - order derivative still hold.
- **Theorem 1.4**: On a closed manifold, if there exists a weak subsolution that satisfies the condition at the tangent cone at infinity, then the a priori estimate of the second - order derivative holds.
- **Proposition 1.5**: On the type - 1 cone \( \Gamma \), the condition at the tangent cone at infinity and the \(\mathcal{C}\)-subsolution condition are equivalent.
### Conclusion
Through these results, the author not only extends the scope of application of the existing theory but also provides new directions for future research, especially regarding how to handle a priori estimates and solvability of fully nonlinear elliptic equations under more general conditions.