Yes, if they are trying to break into your network in a particularly obvious way, but no for the general method of attacking WPA2 with a weak pre-shared key.
If they are trying to brute force into your network by repeated attempted logins (e.g., exploit the flaws of WPS - Wifi "Protected" Setup ), you can listen in on the network traffic, you would see the large fraction of unsuccessful login attempts. (Or even easier your wireless AP logs failed wifi login attempts).
This would require setting up wireshark on a computer that has a wifi card that can listen in promiscuous mode to the traffic. I've heard that this cannot be done in windows (not a windows user so I am not sure of this), so you may have to use linux or a Mac.
Granted most attacks on WPA2 do not work like this (excluding those on WPS). Most attacks work by observing someone else's successful WPA2 handshake (listening in promiscuous mode) to a given access point (knowing the BSSID). Then locally run take that handshake through a password cracking program until they find a match. This cracking is done entirely by eavesdropping and offline cracking, so would not be detectable.
However, these attacks are computationally expensive and will only work in practice against weak passwords or slightly stronger passwords with a common SSID (like NETGEAR / linksys / default / dlink / wireless / Home). This is because the actual password used is a combination of the SSID + preshared key. Rainbow tables (precomputed tables) are generated for common SSID's, and when used, increase the speed of recovering your password. It is advised to change your networks SSID to something unique.
If you have a passphrase that is high entropy (not likely to be in any dictionary attack) and a unique ssid, these offline attacks against your network will be unsuccessful.