You can use a custom tick formatter from the dates package to render the dates as you wish.
Extending your code example:
import numpy as np
import datetime
import matplotlib.pyplot as plt
from matplotlib import dates
data = np.random.random(1440,)
# I can represent minutes as integers:
mins = np.arange(1440,dtype=np.int)
# convert to datetime
times=np.array([datetime.datetime(2014, 5, 13, int(p/60), p%60) for p in mins])
# and plot for every 20 samples:
plt.plot(times[1::20], data[1::20])
# generate a formatter, using the fields required
fmtr = dates.DateFormatter("%H:%M")
# need a handle to the current axes to manipulate it
ax = plt.gca()
# set this formatter to the axis
ax.xaxis.set_major_formatter(fmtr)
plt.show()
The format string is defined as per the strftime
docs.